The field of artificial intelligence is built on object detection techniques. YOU ONLY LOOK
ONCE (YOLO) algorithm and it's more evolved versions are briefly described in this research survey. This
survey is all about YOLO and convolution neural networks (CNN) in the direction of real time object detection.
YOLO does generalized object representation more effectively without precision losses than other object
detection models. CNN architecture models have the ability to eliminate highlights and identify objects in any
given image. When implemented appropriately, CNN models can address issues like deformity diagnosis,
creating educational or instructive application, etc. This article reached at number of observations and
perspective findings through the analysis. Also it provides support for the focused visual information and
feature extraction in the financial and other industries, highlights the method of target detection and feature
selection, and briefly describes the development process of yolo algorithm
Yinyin Liu presents a model for object detection and localization, called Fast-RCNN. She will show how to introduce a ROI pooling layer into neon, and how to add the PASCAL VOC dataset to interface with model training and inference. Lastly, Yinyin will run through a demo on how to apply the trained model to detect new objects.
You Only Look Once: Unified, Real-Time Object DetectionDADAJONJURAKUZIEV
YOLO, a new approach to object detection. A single neural network predicts bounding boxes and class probabilities directly from full images in one evaluation.
Object detection is a computer technology related to computer vision and image processing that deals with detecting instances of semantic objects of a certain class (such as humans, buildings, or cars) in digital images and videos. Well-researched domains of object detection include face detection and pedestrian detection. Object detection has applications in many areas of computer vision, including image retrieval and video surveillance.
Slides from the UPC reading group on computer vision about the following paper:
Redmon, Joseph, Santosh Divvala, Ross Girshick, and Ali Farhadi. "You only look once: Unified, real-time object detection." arXiv preprint arXiv:1506.02640 (2015).
Using a public dataset of images of maritime vessels provided by Analytics Vidhya, manual annotations were made on a subsample of images with Roboflow using the ground truth classifications provided by the dataset. YOLOv5, a prominent open source family of object detection models that comes with an out-of-the-box pre-training on the Common Objects in Context (COCO) dataset, was used to train on annotations of subclassifications of maritime vessels. YOLOv5 provides significant results in detecting a boat. The training, validation, and test set of images trained YOLOv5 in the cloud using Google Colab. Three of our five subclasses, namely, cruise ships, ROROs (Roll On Roll Off, typically car carriers), and military ships, have very distinct shapes and features and yielded positive results. Two of our subclasses, namely, the tanker and cargo ship, have similar characteristics when the cargo ship is unloaded and not carrying any cargo containers. This yielded interesting misclassifications that could be improved in future work. Our trained model resulted in the validation metric of mean Average Precision (mAP@.5) of 0.932 across all subclassification of ships.
Yinyin Liu presents a model for object detection and localization, called Fast-RCNN. She will show how to introduce a ROI pooling layer into neon, and how to add the PASCAL VOC dataset to interface with model training and inference. Lastly, Yinyin will run through a demo on how to apply the trained model to detect new objects.
You Only Look Once: Unified, Real-Time Object DetectionDADAJONJURAKUZIEV
YOLO, a new approach to object detection. A single neural network predicts bounding boxes and class probabilities directly from full images in one evaluation.
Object detection is a computer technology related to computer vision and image processing that deals with detecting instances of semantic objects of a certain class (such as humans, buildings, or cars) in digital images and videos. Well-researched domains of object detection include face detection and pedestrian detection. Object detection has applications in many areas of computer vision, including image retrieval and video surveillance.
Slides from the UPC reading group on computer vision about the following paper:
Redmon, Joseph, Santosh Divvala, Ross Girshick, and Ali Farhadi. "You only look once: Unified, real-time object detection." arXiv preprint arXiv:1506.02640 (2015).
Using a public dataset of images of maritime vessels provided by Analytics Vidhya, manual annotations were made on a subsample of images with Roboflow using the ground truth classifications provided by the dataset. YOLOv5, a prominent open source family of object detection models that comes with an out-of-the-box pre-training on the Common Objects in Context (COCO) dataset, was used to train on annotations of subclassifications of maritime vessels. YOLOv5 provides significant results in detecting a boat. The training, validation, and test set of images trained YOLOv5 in the cloud using Google Colab. Three of our five subclasses, namely, cruise ships, ROROs (Roll On Roll Off, typically car carriers), and military ships, have very distinct shapes and features and yielded positive results. Two of our subclasses, namely, the tanker and cargo ship, have similar characteristics when the cargo ship is unloaded and not carrying any cargo containers. This yielded interesting misclassifications that could be improved in future work. Our trained model resulted in the validation metric of mean Average Precision (mAP@.5) of 0.932 across all subclassification of ships.
PR-207: YOLOv3: An Incremental ImprovementJinwon Lee
TensorFlow Korea 논문읽기모임 PR12 207번째 논문 review입니다
이번 논문은 YOLO v3입니다.
매우 유명한 논문이라서 크게 부연설명이 필요없을 것 같은데요, Object Detection algorithm들 중에 YOLO는 굉장히 특색있는 one-stage algorithm입니다. 이 논문에서는 YOLO v2(YOLO9000) 이후에 성능 향상을 위하여 어떤 것들을 적용하였는지 하나씩 설명해주고 있습니다. 또한 MS COCO의 metric인 average mAP에 대해서 비판하면서 mAP를 평가하는 방법에 대해서도 얘기를 하고 있는데요, 자세한 내용은 영상을 참고해주세요~
논문링크: https://arxiv.org/abs/1804.02767
영상링크: https://youtu.be/HMgcvgRrDcA
Object detection is an important computer vision technique with applications in several domains such as autonomous driving, personal and industrial robotics. The below slides cover the history of object detection from before deep learning until recent research. The slides aim to cover the history and future directions of object detection, as well as some guidelines for how to choose which type of object detector to use for your own project.
Object Detection using Deep Neural NetworksUsman Qayyum
Recent Talk at PI school covering following contents
Object Detection
Recent Architecture of Deep NN for Object Detection
Object Detection on Embedded Computers (or for edge computing)
SqueezeNet for embedded computing
TinySSD (object detection for edge computing)
Slide for Multi Object Tracking by Md. Minhazul Haque, Rajshahi University of Engineering and Technology
* Object
* Object Tracking
* Application
* Background Study
* How it works
* Multi-Object Tracking
* Solution
* Future Works
In Comparison with other object detection algorithms, YOLO proposes the use of an end-to-end neural network that makes predictions of bounding boxes and class probabilities all at once.
This is a presentation on the Yolo(You Only Look Once) object detection system. This is a state-of-the-art system that is works very fast. The presentation has been derived from the paper cited below
@article{yolov3,
title={YOLOv3: An Incremental Improvement},
author={Redmon, Joseph and Farhadi, Ali},
journal = {arXiv},
year={2018}
}
Overview Of Video Object Tracking SystemEditor IJMTER
The goal of video object tracking system is segmenting a region of interest from a video
scene and keeping track of its motion, positioning and occlusion. There are the three steps of video
object tracking system those are object detection, object classification and object tracking. Object
detection is performed to check existence of objects in video. Then the detected object can be
classified in various categories on the basis on their shape, motion, color and texture. Object tracking
is performed using monitoring object changes. This paper we are going to take overview of different
object detection, object classification and object tracking techniques and also the comparison of
different techniques used for various stages of tracking.
Object detection is a main role in image processing.the proposed methods detect various multiple object detection using image processing so provide a really to solving the security problem.
Computer vision has received great attention over the last two decades.
This research field is important not only in security-related software but also in the advanced interface between people and computers, advanced control methods, and many other areas.
PR-207: YOLOv3: An Incremental ImprovementJinwon Lee
TensorFlow Korea 논문읽기모임 PR12 207번째 논문 review입니다
이번 논문은 YOLO v3입니다.
매우 유명한 논문이라서 크게 부연설명이 필요없을 것 같은데요, Object Detection algorithm들 중에 YOLO는 굉장히 특색있는 one-stage algorithm입니다. 이 논문에서는 YOLO v2(YOLO9000) 이후에 성능 향상을 위하여 어떤 것들을 적용하였는지 하나씩 설명해주고 있습니다. 또한 MS COCO의 metric인 average mAP에 대해서 비판하면서 mAP를 평가하는 방법에 대해서도 얘기를 하고 있는데요, 자세한 내용은 영상을 참고해주세요~
논문링크: https://arxiv.org/abs/1804.02767
영상링크: https://youtu.be/HMgcvgRrDcA
Object detection is an important computer vision technique with applications in several domains such as autonomous driving, personal and industrial robotics. The below slides cover the history of object detection from before deep learning until recent research. The slides aim to cover the history and future directions of object detection, as well as some guidelines for how to choose which type of object detector to use for your own project.
Object Detection using Deep Neural NetworksUsman Qayyum
Recent Talk at PI school covering following contents
Object Detection
Recent Architecture of Deep NN for Object Detection
Object Detection on Embedded Computers (or for edge computing)
SqueezeNet for embedded computing
TinySSD (object detection for edge computing)
Slide for Multi Object Tracking by Md. Minhazul Haque, Rajshahi University of Engineering and Technology
* Object
* Object Tracking
* Application
* Background Study
* How it works
* Multi-Object Tracking
* Solution
* Future Works
In Comparison with other object detection algorithms, YOLO proposes the use of an end-to-end neural network that makes predictions of bounding boxes and class probabilities all at once.
This is a presentation on the Yolo(You Only Look Once) object detection system. This is a state-of-the-art system that is works very fast. The presentation has been derived from the paper cited below
@article{yolov3,
title={YOLOv3: An Incremental Improvement},
author={Redmon, Joseph and Farhadi, Ali},
journal = {arXiv},
year={2018}
}
Overview Of Video Object Tracking SystemEditor IJMTER
The goal of video object tracking system is segmenting a region of interest from a video
scene and keeping track of its motion, positioning and occlusion. There are the three steps of video
object tracking system those are object detection, object classification and object tracking. Object
detection is performed to check existence of objects in video. Then the detected object can be
classified in various categories on the basis on their shape, motion, color and texture. Object tracking
is performed using monitoring object changes. This paper we are going to take overview of different
object detection, object classification and object tracking techniques and also the comparison of
different techniques used for various stages of tracking.
Object detection is a main role in image processing.the proposed methods detect various multiple object detection using image processing so provide a really to solving the security problem.
Computer vision has received great attention over the last two decades.
This research field is important not only in security-related software but also in the advanced interface between people and computers, advanced control methods, and many other areas.
An assistive model of obstacle detection based on deep learning: YOLOv3 for v...IJECEIAES
The World Health Organization (WHO) reported in 2019 that at least 2.2 billion people were visual-impairment or blindness. The main problem of living for visually impaired people have been facing difficulties in moving even indoor or outdoor situations. Therefore, their lives are not safe and harmful. In this paper, we proposed an assistive application model based on deep learning: YOLOv3 with a Darknet-53 base network for visually impaired people on a smartphone. The Pascal VOC2007 and Pascal VOC2012 were used for the training set and used Pascal VOC2007 test set for validation. The assistive model was installed on a smartphone with an eSpeak synthesizer which generates the audio output to the user. The experimental result showed a high speed and also high detection accuracy. The proposed application with the help of technology will be an effective way to assist visually impaired people to interact with the surrounding environment in their daily life.
DEEP LEARNING APPROACH FOR EVENT MONITORING SYSTEMIJMIT JOURNAL
With an increasing number of extreme events and complexity, more alarms are being used to monitor
control rooms. Operators in the control rooms need to monitor and analyze these alarms to take suitable
actions to ensure the system’s stability and security. Security is the biggest concern in the modern world. It
is important to have a rigid surveillance that should guarantee protection from any sought of hazard.
Considering security, Closed Circuit TV (CCTV) cameras are being utilized for reconnaissance, but these
CCTV cameras require a person for supervision. As a human being, there can be a possibility to be tired
off in supervision at any point of time. So, we need a system to detect automatically. Thus, we came up with
a solution using YOLO V5. We have taken a data set and used robo-flow framework to enhance the existing
images into numerous variations where it will create a copy of grey scale image, a copy of its rotation and
a copy of its blurred version which will be used to get an enlarged data set. This work mainly focuses on
providing a secure environment using CCTV live footage as a source to detect the weapons. Using YOLO
algorithm, it divides an image from the video into grid system and each grid detects an object within itself
You only look once model-based object identification in computer visionIAESIJAI
You only look once version 4 (YOLOv4) is a deep-learning object detection algorithm. It is used to decrease parameters and simplify network structures, making it suited for mobile and embedded device development. The YOLO detector can foresee an object's Class, bounding box, and probability of that Object's Class being found inside that bounding box. A probability value for each bounding box represents the likelihood of a given item class in that bounding box. Global features, channel attention, and special attention are also applied to extract more compelling information. Finally, the model combines the auxiliary and backbone networks to create the YOLOv4's entire network topology. Using custom functions developed upon YOLOv4, we get the count of the objects and a crop around the objects detected with a confidence score that specifies the probability of the thing seen being the same Class as predicted by YOLOv4. A confidence threshold is implemented to eliminate the detections with low confidence.
Deep-learning based single object tracker for night surveillance IJECEIAES
Tracking an object in night surveillance video is a challenging task as the quality of the captured image is normally poor with low brightness and contrast. The task becomes harder for a small object as fewer features are apparent. Traditional approach is based on improving the image quality before tracking is performed. In this paper, a single object tracking algorithm based on deep-learning approach is proposed to exploit its outstanding capability of modelling object’s appearance even during night. The algorithm uses pre-trained convolutional neural networks coupled with fully connected layers, which are trained online during the tracking so that it is able to cater for appearance changes as the object moves around. Various learning hyperparameters for the optimization function, learning rate and ratio of training samples are tested to find optimal setup for tracking in night scenarios. Fourteen night surveillance videos are collected for validation purpose, which are captured from three viewing angles. The results show that the best accuracy is obtained by using Adam optimizer with learning rate of 0.00075 and sampling ratio of 2:1 for positive and negative training data. This algorithm is suitable to be implemented in higher level surveillance applications such as abnormal behavioral recognition.
Similar to Real Time Object Detection System with YOLO and CNN Models: A Review (20)
Custom Hardware and Software Integration: Bluetooth Based Wireless Thermal P...Springer
In this paper, the communication between
Arduino Mega microcontroller and mini thermal printer is
implemented by hardware and software integration. The
microcontroller is programmed to transfer the data from
microcontroller to printer using a serial communication
protocol such as universal asynchronous receiver and
transmitter (UART) via Serial communication devices such as
MAX232 converter, DB9 connector which is nothing but RS232
serial port, and universal serial bus (USB) communication port.
Data sent by the microcontroller is made to print from the
printer and it is shown in the result section. Digital pins '3' and
'4' on the microcontroller are used as virtual Rx and Tx serial
lines while leaving the main serial port open for debugging
purposes. Interfacing between UART of microcontroller,
MAX232, serial port, and the USB port of printer are well
established with the complete schematic diagram. The mini
thermal printer prints whatever is sent by the microcontroller.
A simple code similar to the one used for the serial monitor
works for the printer. The baud rate needs to be set at 9600 for
the microcontroller to communicate with the printer. The
system is designed to print data wirelessly using Bluetooth HC05 for restaurant and hospital management applications. The
system has high adaptability. It can be used in many situations
not only for restaurant and hospital management.
Implementation of IoT in Agriculture: A Scientific Approach for Smart Irriga...Springer
Digital technologies empower the
transformation into data-driven, intelligent, agile, and
autonomous farm operations and are typically considered a
key to addressing the grand challenges in agriculture. To avoid
unscientific water supply for plantation as well as to save the
water and also yield the better crop, therefore, to increase
production efficiency out of smart irrigation and to send the
status of irrigation at standard environmental conditions, The
Internet of Things (IoT) based prototype is designed and
implemented. The prototype automatically turns ON /OFF the
motor pump based on the moisture level of the soil by taking
the temperature and humidity of the environment near the
plantation into consideration (In India, the standard
parameters for watering the vegetable plantation are
Humidity>60%, Temperature < 25°C and Humidity<40% ).
The prototype is designed with an ESP32S microcontroller
with DHT 11 and a moisture sensor. Arduino IDE development
tool is used for programming ESP32S using embedded C
programming language. The prototype is configured,
programmed, and connected to the Arduino IoT cloud. The
data of temperature, humidity, and moisture are received via
message queuing telemetry transport (MQTT) protocol on IoT
cloud through public IP therefore the data can be accessed
worldwide. The authorized person can access the data and
control the motor pump from anywhere across the world. The
test data obtained out of the prototype over the cloud and at
the system are presented in the result section.
Implementation of Tiny Machine Learning Models on Arduino 33 – BLE for Gestur...Springer
In this article gesture recognition and speech recognition applications are implemented on embedded systems with Tiny Machine Learning (TinyML). The main benefit of using TinyML is its portability. It can be run on cheap microcontrollers with tiny batteries and low power consumption, and it can easily integrate machine learning with virtually anything. It also has the benefit of increased security due to local nature of computing. The benefit of using Arduino Nano 33 BLE sense is that it has a set of sensors embedded on the top, which gives us a lot of options to try ideas without having to generate the circuit to such sensors in prototyping board. It features 3-axis accelerometer, 3-axis gyroscope and 3-axis magnetometer.
The gesture recognition, provides an innovative approach nonverbal communication. It has wide applications in human-computer interaction and sign language. Here in the implementation of hand gesture recognition, TinyML model is trained and deployed from EdgeImpulse framework for hand gesture recognition and based on the hand movements, Arduino Nano 33 BLE device having 6-axis IMU can find out the direction of movement of hand. The Speech is a mode of communication. Speech recognition is a way by which the statements or commands of human speech is understood by the computer which reacts accordingly. The main aim of speech recognition is to achieve communication between man and machine. Here in the implementation of speech recognition, the TinyML model is trained and deployed from the EdgeImpulse framework for speech recognition and based on the keyword pronounced by human, Arduino Nano 33
Bidirectional DC-DC converter circuits and smart control algorithms_ a review...Springer
The entire article has been dedicated to cover the current state of the art in bidirectional DC-DC converter topologies and its smart control algorithms, identified the research gaps and concluded with the motivation for taking up the work. It covers the literature survey of bidirectional buck–boost DC-DC converters, and control schemes are carried out on two aspects, one is on topology perspective and another one is on control schemes. Different topologies with and without transformers of bidirectional DC-DC converters are discussed. Non-isolated converters establish the DC path between input and output sides while transformer-based converters cancel the DC path in between input and output sides since it introduces AC line between two DC lines just like in flyback converter. Transformer-less converter is preferred when there is no much protection needed for load from high voltage levels, also these converters are used in high-power applications. The bidirectional DC-DC converter can switch the power between two DC sources and the load. To do so, it has to use proper control schemes and control algorithms. It can store the excess energy in batteries or in super capacitors. In contrast, isolated topologies contain transformers in their circuits. Due to this, it offers advantages like safeguarding sensitive loads from high power which is at input side. In addition to it, multiple input and output ports can be established. With the isolation in DC-DC converters, input and output sections are separated from electrical stand point of view. With isolation, both input and output sections will not be having common ground point. The DC path is removed with isolation due to usage of transformer in DC-DC converters. In contrast to its features, it is capable to be used in low-power applications since transformer is switching at high frequency, the size of the coil reduces and hence it can handle limited rate of current. The bidirectional DC-DC converters are categorized based on isolation property so-called isolated bidirectional converters. Features and applications of each topology are presented. Comparative analysis w.r.t research gaps between all the topologies is presented. Also the scope of control schemes with artificial intelligence is discussed. Pros and cons of each control scheme, i.e. research gaps in control schemes and impact of control scheme for bidirectional DC-DC converters, are also presented.
Microcontroller based bidirectional buck–boost converter for photo-voltaic po...Springer
A common configuration for a stand-alone PV power system may consist of three converters: a buck converter for the PV panel
to charge the battery, a boost converter for the battery to discharge to the load and one for the load voltage regulation. Such a system
requires a coordinated control scheme for three converters which can be complicated. A simple structure for a stand-alone PV plant
consists of a PV array, a battery unit, and its associated bidirectional converter which is a combination of a buck and boost converter.
When controlled properly the system can provide uninterrupted power to the load, despite the intermittent availability of sunlight. In
this paper complete design of the converter is carried out and the simulation has been performed using Psim. From the simulation,
the graphs are presented to show the converter working in buck mode and boost mode. Controller is designed to take care of mode
transition, buck to boost and boost to buck mode automatically based on source voltage. Hardware implementation has been done
using microcontroller (8051).
Stability and Dynamic Response of Analog and Digital Control loops of Bidire...Springer
This article presents the stability and dynamic
response of open loop and closed loop control of Bidirectional
buck-boost converter(BBC) using PID and PIDN controllers
through transfer function model implemented in MATLAB code.
In order to ensure the stability of switch mode power supplies the
control loop behaviors need to be characterized. Improvement of
stability of BBC using PID/PIDN compensators is demonstrated
in both analog and digital domains by plotting bode plots. Step
response of BBC using PID /PIDN controllers are plotted that
defines the dynamic behavior of the system.PIDN compensator
is proposed to maintain a healthy balance between the stability
and transient behavior since both are indirectly proportional.
CFD Simulation of By-pass Flow in a HRSG module by R&R Consult.pptxR&R Consult
CFD analysis is incredibly effective at solving mysteries and improving the performance of complex systems!
Here's a great example: At a large natural gas-fired power plant, where they use waste heat to generate steam and energy, they were puzzled that their boiler wasn't producing as much steam as expected.
R&R and Tetra Engineering Group Inc. were asked to solve the issue with reduced steam production.
An inspection had shown that a significant amount of hot flue gas was bypassing the boiler tubes, where the heat was supposed to be transferred.
R&R Consult conducted a CFD analysis, which revealed that 6.3% of the flue gas was bypassing the boiler tubes without transferring heat. The analysis also showed that the flue gas was instead being directed along the sides of the boiler and between the modules that were supposed to capture the heat. This was the cause of the reduced performance.
Based on our results, Tetra Engineering installed covering plates to reduce the bypass flow. This improved the boiler's performance and increased electricity production.
It is always satisfying when we can help solve complex challenges like this. Do your systems also need a check-up or optimization? Give us a call!
Work done in cooperation with James Malloy and David Moelling from Tetra Engineering.
More examples of our work https://www.r-r-consult.dk/en/cases-en/
NO1 Uk best vashikaran specialist in delhi vashikaran baba near me online vas...Amil Baba Dawood bangali
Contact with Dawood Bhai Just call on +92322-6382012 and we'll help you. We'll solve all your problems within 12 to 24 hours and with 101% guarantee and with astrology systematic. If you want to take any personal or professional advice then also you can call us on +92322-6382012 , ONLINE LOVE PROBLEM & Other all types of Daily Life Problem's.Then CALL or WHATSAPP us on +92322-6382012 and Get all these problems solutions here by Amil Baba DAWOOD BANGALI
#vashikaranspecialist #astrologer #palmistry #amliyaat #taweez #manpasandshadi #horoscope #spiritual #lovelife #lovespell #marriagespell#aamilbabainpakistan #amilbabainkarachi #powerfullblackmagicspell #kalajadumantarspecialist #realamilbaba #AmilbabainPakistan #astrologerincanada #astrologerindubai #lovespellsmaster #kalajaduspecialist #lovespellsthatwork #aamilbabainlahore#blackmagicformarriage #aamilbaba #kalajadu #kalailam #taweez #wazifaexpert #jadumantar #vashikaranspecialist #astrologer #palmistry #amliyaat #taweez #manpasandshadi #horoscope #spiritual #lovelife #lovespell #marriagespell#aamilbabainpakistan #amilbabainkarachi #powerfullblackmagicspell #kalajadumantarspecialist #realamilbaba #AmilbabainPakistan #astrologerincanada #astrologerindubai #lovespellsmaster #kalajaduspecialist #lovespellsthatwork #aamilbabainlahore #blackmagicforlove #blackmagicformarriage #aamilbaba #kalajadu #kalailam #taweez #wazifaexpert #jadumantar #vashikaranspecialist #astrologer #palmistry #amliyaat #taweez #manpasandshadi #horoscope #spiritual #lovelife #lovespell #marriagespell#aamilbabainpakistan #amilbabainkarachi #powerfullblackmagicspell #kalajadumantarspecialist #realamilbaba #AmilbabainPakistan #astrologerincanada #astrologerindubai #lovespellsmaster #kalajaduspecialist #lovespellsthatwork #aamilbabainlahore #Amilbabainuk #amilbabainspain #amilbabaindubai #Amilbabainnorway #amilbabainkrachi #amilbabainlahore #amilbabaingujranwalan #amilbabainislamabad
Final project report on grocery store management system..pdfKamal Acharya
In today’s fast-changing business environment, it’s extremely important to be able to respond to client needs in the most effective and timely manner. If your customers wish to see your business online and have instant access to your products or services.
Online Grocery Store is an e-commerce website, which retails various grocery products. This project allows viewing various products available enables registered users to purchase desired products instantly using Paytm, UPI payment processor (Instant Pay) and also can place order by using Cash on Delivery (Pay Later) option. This project provides an easy access to Administrators and Managers to view orders placed using Pay Later and Instant Pay options.
In order to develop an e-commerce website, a number of Technologies must be studied and understood. These include multi-tiered architecture, server and client-side scripting techniques, implementation technologies, programming language (such as PHP, HTML, CSS, JavaScript) and MySQL relational databases. This is a project with the objective to develop a basic website where a consumer is provided with a shopping cart website and also to know about the technologies used to develop such a website.
This document will discuss each of the underlying technologies to create and implement an e- commerce website.
Hierarchical Digital Twin of a Naval Power SystemKerry Sado
A hierarchical digital twin of a Naval DC power system has been developed and experimentally verified. Similar to other state-of-the-art digital twins, this technology creates a digital replica of the physical system executed in real-time or faster, which can modify hardware controls. However, its advantage stems from distributing computational efforts by utilizing a hierarchical structure composed of lower-level digital twin blocks and a higher-level system digital twin. Each digital twin block is associated with a physical subsystem of the hardware and communicates with a singular system digital twin, which creates a system-level response. By extracting information from each level of the hierarchy, power system controls of the hardware were reconfigured autonomously. This hierarchical digital twin development offers several advantages over other digital twins, particularly in the field of naval power systems. The hierarchical structure allows for greater computational efficiency and scalability while the ability to autonomously reconfigure hardware controls offers increased flexibility and responsiveness. The hierarchical decomposition and models utilized were well aligned with the physical twin, as indicated by the maximum deviations between the developed digital twin hierarchy and the hardware.
Overview of the fundamental roles in Hydropower generation and the components involved in wider Electrical Engineering.
This paper presents the design and construction of hydroelectric dams from the hydrologist’s survey of the valley before construction, all aspects and involved disciplines, fluid dynamics, structural engineering, generation and mains frequency regulation to the very transmission of power through the network in the United Kingdom.
Author: Robbie Edward Sayers
Collaborators and co editors: Charlie Sims and Connor Healey.
(C) 2024 Robbie E. Sayers
Welcome to WIPAC Monthly the magazine brought to you by the LinkedIn Group Water Industry Process Automation & Control.
In this month's edition, along with this month's industry news to celebrate the 13 years since the group was created we have articles including
A case study of the used of Advanced Process Control at the Wastewater Treatment works at Lleida in Spain
A look back on an article on smart wastewater networks in order to see how the industry has measured up in the interim around the adoption of Digital Transformation in the Water Industry.
Real Time Object Detection System with YOLO and CNN Models: A Review
1. Real Time Object Detection System with
YOLO and CNN Models: A Review
Viswanatha V1*
Asst.Professor, Electronics and Communication Engineering Department,
Nitte Meenakshi Institute of Technology, Bangalore.
Email – viswas779@gmail.com
Chandana R K2
MTech student, Electronics and Communication Engineering Department
Nitte Meenakshi Institute of Technology, Bangalore. Email- chandanark99@gmail.com
Ramachandra A.C3
Professor and Head, Electronics and Communication Engineering Department
Nitte Meenakshi Institute of Technology, Bangalore. Email- ramachandra.ac@nmit.ac.in
Abstract- The field of artificial intelligence is built on object detection techniques. YOU ONLY LOOK
ONCE (YOLO) algorithm and it's more evolved versions are briefly described in this research survey. This
survey is all about YOLO and convolution neural networks (CNN) in the direction of real time object detection.
YOLO does generalized object representation more effectively without precision losses than other object
detection models. CNN architecture models have the ability to eliminate highlights and identify objects in any
given image. When implemented appropriately, CNN models can address issues like deformity diagnosis,
creating educational or instructive application, etc. This article reached at number of observations and
perspective findings through the analysis. Also it provides support for the focused visual information and
feature extraction in the financial and other industries, highlights the method of target detection and feature
selection, and briefly describe the development process of yolo algorithm.
Keywords- Object Detection, YOLO, CNN
I. INTRODUCTION
Object detection strategies are the establishment for the engineered insights subject. This paper offers a brief
evaluation of the you only look once (YOLO) algorithm of rules and another prevalent variation. Through the
assessment, it is understood that numerous comments and proper results show the variations and likenesses
between Yolo version and between Yolo and convolutional neural networks (CNNS). The relevant perception is
the Yolo algorithm development continues to be Ongoing. This article in brief depicts the improvement procedure
of the Yolo set of rules, summarizes the methods of goal Recognition and characteristic choice, and provides
literature assistance for the centered image news and characteristic extraction as shown in fig.1 [1].
Journal of Xi'an University of Architecture & Technology
Volume XIV, Issue 7, 2022
ISSN No : 1006-7930
Page No: 144
2. Figure.1
YOLO is a new approach of object detection. The classifiers are used to detection in earlier works in an object
detection. It is consider here that frame object detection as a regression issue to spatially separated bounding
package container and associated class probabilities as an alternative object detection as shown in fig.2.
Figure.2
In a single evaluation, a single neural network predicts a boundary bins and class possibilities directly from
complete images. The entire direction pipeline may be optimized start to finish because it is a single network. On
the effectiveness of detection without delay a unified architecture moves incredibly quickly and as well 45 frames
per 2D frame yolo model process image in real time. Yolo, a scaled down version of the network, achieves double
the coverage of the other real time detectors while operating at an incredible 125 frames per second. Yolo makes
more localization errors than modern detection algorithm but is less likely to anticipate false positives based on
past data. Yolo eventually picks up and highly preferred representation of items while generalizing from natural
photographs to other domain names like painting, its outperform competing detection techniques like DPM and
R-CNN [2].
II. Approaches of implementation of YOLO and CNN
Object detection is a technology that detects the semantic Objects of a category in virtual snap shots and films.
Certainly, one of its Actual-time packages is self-riding automobiles. In this, our challenge is to locate a couple
of items from a photo. The maximum common Item to come across on this utility is the car, motorcycle, and
Pedestrian. For locating the gadgets within the photograph, Object localization and should find multiple item in
real-time structures. There are various techniques for item Detection, they can be break up into classes, first is the
Algorithms primarily based on classifications. CNN and RNN come below this category. On this, pick out the
involved Regions from the picture and ought to classify them the use of Convolutional neural community. This
technique may be very sluggish Due to the fact it should run a prediction for every decided on Vicinity. The
second one class is the algorithms primarily based on Regressions. Yolo approach comes below this category. In
This, it might not choose the fascinated regions from the photograph. Instead, it expect the training and bounding
containers of the complete picture at a single run of the algorithm and detect a couple of gadgets using an
unmarried neural community. Yolo Set of rules is rapid as compared to other classification Algorithms. In actual
time our algorithm technique 45 frames consistent with 2d. Yolo algorithm makes localization mistakes however
Predicts less fake positives in the background [3].
Humans can without problems come across and pick out gadgets of their environment without attention in their
instances, irrespective of what position they're in and whether or not they are the wrong way up, one-of-a-kind in
shade or texture, in part occluded, and so forth. To extract information about the objects and shapes in a picture,
the same item detection and recognition on a computer requires a lot of processing. Identifying an object in an
image or video is referred to end as an object detection in computer vision. The main steps in object detection or
future extraction, which is imported for surveillance, cancer reduction, car identification, and underwater object
detection, among other applications. Different approaches had been used to accurately and correctly identify the
object for specialized packages. These suggested solutions still have an issue with inaccuracy and inefficiency,
though. Device learning and deep neural network approaches are more successful in addressing these issues of
object detection [4].
Convolution neural networks CNN for widely used in image processing and other areas with the development of
artificial intelligence because of their fantastic performance. However, because it's a set of rules is so complex
computationally, CNN faces enormous difficulties capturing the interest of mobile devices. FPGA becoming ideal
choices for CNN deployment due to their advantages of high performance, reprogramming, and seldom energy
usage. The yolo algorithm views target detection as a regression problem in comparison to other CNN algorithms.
Journal of Xi'an University of Architecture & Technology
Volume XIV, Issue 7, 2022
ISSN No : 1006-7930
Page No: 145
3. It is a one-step algorithm that runs quickly and requires little calculation. It is much more appropriate for FPGA
hardware structure. The issue of massive computational of CNN and constrained assets on FPGA chip is resolved,
and the parallelism of FPGA is used to boost the CNN. This is accomplished by improving the Yolo network
version and fixed point, among other things. Experimental finding demonstrate that the strategy suggested in this
paper as significantly reduced the operating cost while maintaining accuracy and as an essential reasonable cost
within the creation of mobile terminals and real time computing. In order to speed up the deep learning yolo
community, this paper offers an advanced set of guidelines for deep learning Yolo network that are entirely based
on Xilinx FPGA. By using the parallelism function of the FPGA to speed up the CNN, the acceleration set of rules
specifically addresses the issue that the CNN requires a significant amount of calculation while the FPGA has
limited resources on the chip. The implementation of this set of rules specifically includes all three parts: the first
one is the FPGA Yolo community design the second one is the selection of the activation characteristics and the
last one is 16 bit constant point optimization of the weight parameter [5].
Development plan is presented as a result of the lack of common accuracy and miss detection in the way of actual
roots in gold detection through the Yolo V3 community in order to investigate the anchor range and component
ratio, the original network clustering set of rules is updated using the ok-means clustering algorithm similarly, so
that you can enhance the performance of the road target detection set of rules, the present network output is
upgraded, and a 104*104 feature detection layer is brought, and the feature map output through 8 instances
sampling can be output through 2 instances up sampling, and 4 the characteristic maps of down-sampling are
stitched together, and the 104*104 feature maps received can efficiently lessen the disappearance of features. Via
the experimental effects, it is able to see that compared with the stepped forward yolov3 set of rules, the common
detection accuracy of the stepped forward set of rules for road goal detection is quite excessive to a few.17%, and
the missing detection rate is reduced by 5.62% [6].
In this paper author proposed laptop vision functions of onboard cameras and embedded devices are widely
employed with general unmanned aerial vehicles (UAV). However ,it is very difficult program to study the real
time seen through the object identification approach go to the limited memory and computing capability of
embedded devices at the UAV platform. This study evaluates the performance of different you yolo collection
methods on the Pascal voc dataset, employees map and frame per second at assessment measures, and applies the
learning outcomes to the xtdrone UAV simulation platform for testing in order to overcome those difficulties. The
map of Yolo V4 is 87.48%, which is 14.2% better than that of yolo V3 according to our evaluation of Yolo V3,
Yolo V3 tiny, and yoov3-spp3 and yolov4-tiny on the pascal voc benchmark dataset. Frame per second reaches
72, and the test sets check time is one or 103.8 seconds, the validation sets shortest test time is Yolo V4, but the
map only achieves 50.06% accuracy which is less precise than Yolo V3 tiny. Using simulation on the xtdrone
platform to analyze the performance of five modules on the Pascal voc data set come on this paper ultimately
comes to the conclusion that Yolo V3 can match the demands of real time, lightweight, and excessive precision
[7].
The main objective of real time object detection is to locate the location of an object in a supply picture accurately
and mark the item with the appropriate category. In this paper it used actual time object detection you appearance
best as soon as (yolo) set of rules to train our device studying version. Yolo is a clever neural network for doing
object detection in actual time and with the assist of coco dataset the set of rules is trained to perceive one of a
kind items in a specific picture. After training this method locate the object in actual time with ninety% accuracy.
The aim of item detection is to come across all times of gadgets from an acknowledged magnificence,
consisting of humans, motors or faces in an image. Commonly, most effective a small quantity of instances of the
object are gift inside the picture, however there are a very huge range of possible places and scales at which they
could occur and that want to by some means be explored. Each of the detection is mentioned with some form of
pose data. This could be as simple because the region of the object, an area and scale, or the volume of the object
described in terms of a bounding field. In different conditions, the pose facts is extra specific and includes the
parameters of a linear or nonlinear transformation. As an example, a face detector may additionally compute the
places of the eyes, nose and mouth, further to the bounding box of the face [8].
Synthetic intelligence is being tailored by the sector on the grounds that past few years and deep gaining
knowledge of performed a crucial position in it. This paper focuses on in depth learning and how it is used to find
and tune the devices. Deep learning uses algorithms that are impacted by the structure and functions of the brain.
Working with such algorithms as the advantage that performance generally improves with growth in data, as
opposed to conventional learning algorithms whose performance stabilizes regardless of boom in the amount of
data a number of most popular research topics in computer vision application have put real time item monitoring
at the forefront. Despite the impressive progress made at this paper, it is still difficult to effectively and accurately
track the objects in real time at a significant level. The capabilities of images and movies for security and oversight
packages are extracted while defining detection and monitoring algorithm. You only look once (Yolo), location
based fully convolutional neural network (RCNN), and quicker samples of popular object detection algorithm (f-
RCNN). RCNN is more accurate than other algorithms, but yolo outperforms them when speed is prioritized over
accuracy. Yolo treats object detection as a regression issue and provides elegance chances for found images.
Journal of Xi'an University of Architecture & Technology
Volume XIV, Issue 7, 2022
ISSN No : 1006-7930
Page No: 146
4. Real world circumstances drive the need to use the Yolo algorithm to find object entities base. When
pace is taken into account as if performance indicator, Yolo outperforms other object detection algorithms.
Regression is used by the community to predict the bounding containers and classes of entities in a single pass of
algorithm, as opposed traversing the network more than once to identify and output the opportunity of the
elegance. This gives it a real and noticeable advantages over earlier technology in terms of increased frame per
second. When compared to other performances like CNN and its various version this result in an object detector
using Yolo V3 with impressive overall performance metrics, including accuracy, precision, and everything it is
different from other. This enables the builders to use the Yolo V3 for real time monitoring packages in order to
benefit from its performance improvements over its antiquated counterparts [9].
By using photographic records enhancement, it is possible to expand the pool of records without actually adding
any new information. In this study, researchers assessed the effectiveness of image manipulation techniques like
the double phantasm and inversion on the overall performance of the face detection for information enhancement
needs. The study found that the limited amount of data available to create a teaching version of the facts was a
weakness in the facts that were acquired. The goal of this people like research paper is to increase the variety of
the facts such that come up and gives various comparable datasets, it can make prediction with accuracy. Yolo
V3 is used to perform face detection on photos, and accuracy results from the dataset and before including records
are compared.
Object detection as received a lot of attention in the field of computer vision carried out in various areas,
including the clinical field, pastime reputation, security tracking in agencies, and automated manage robots.
Famous item detection techniques, before everything, had been recognized by way of the use of feature extraction
techniques together with histogram of orientated gradients (hog), speeded-up strong features (surf), nearby binary
styles (lbp), and additionally coloration histograms. The system of taking pictures the characteristics of gadgets
that could describe the traits of items is the number one approach in the function extraction technique [10].
A straightforward, ready-to-use, unified object detection model that works with complete images. YOLO also
generalized well to unused spaces utilized in applications that depend on quick, strong object detection. A
degenerative show built for recognizing degraded images like obscured and loud pictures has the model
being prepared with these debased pictures. This model performed superior in terms of detecting
degraded pictures and adapted way better with complex scenes. For discovery shallow person on foot highlights,
a YOLO v2 organize was adjusted by including three Pass through layer to them. The number of detection frames
can reach 25 frames per second, which meets the demands of real-time performance. To recognize
indoor impediments an unused strategy of using profound learning together with a light field camera
was utilized. The strategy recognizes the deterrents and perceives its information.
YOLO connected to vehicle entryway board welding panel lines can distinguish and distinguish patch joint
accurately the calculation can too identify the position of the solder joints and more [11].
Object Detection has received a lot of research attention in recent years because of its tight association with video
analysis and picture interpretation. The foundation of conventional object detection technique is shallow trainable
structures and handmade features. Building intricate ensembles that incorporate several low-level picture features
with high-level context from object detectors and scene classifier can readily stabilize their performance. In order
to solve the issues with traditional architectures, more potent tools that can learn semantic, high-level and deeper
features are being offered as a result of deep learning’s quick development. In terms of network architecture,
training methodology, optimization function, etc., these models behave differently. In this paper, it explore object
detection frameworks based on deep learning. A brief background of deep learning is provided before our review.
It brief history of deep learning and its illustrative tool, the convolutional neural network (CNN), is given before
our review. Then it concentrate on common generic object detection architectures with a few changes and helpful
tips to further enhance detection performance.
This paper also provides overview of a number of specific task, such as salient item detection, face
detection and pedestal detection, as different specific detection task exhibit different characteristics. To evaluate
different approaches and get some insightful results experimental analysis are also offered. In order to provide
direction for future work object identification and pertinent neural network based learning systems, a number of
promising directions and tasks are provided [12].
An aspect of your computer vision that is constantly in use is object detection. Object identification has made
remarkable progress, from the first manual future extractor to the future extractor dominated by deep convolution
network. Researchers started paying more attention to CNN in 2012 when Alexnet won the title image net with
significant benefits (convolutional neural report). CNN has made remarkable success since 2012 in many facets
of computer science, and object vision detection is no exception. As of now, two step end single step CNN-based
object detection network can be loosely split into two categories. The two stage object detection network typically
consists of three stages: The first stage involves creating multiple delimitations for the proposal cans; the second
stage involves performing the object classification. R-CNN, Fast R-CNN, SPPNet,R-CNN faster, FPN, and Mask-
RCNN are some of the representative network for the prediction of position information on the bounding boxes
of the proposition. The single phase object detection network is an end to end model, meaning that only one
Journal of Xi'an University of Architecture & Technology
Volume XIV, Issue 7, 2022
ISSN No : 1006-7930
Page No: 147
5. network inference is required from the input image to the network output. Its representative objectives include the
YOLO series, SSD and RetinaNet. The benefits and drawbacks of the two strategies mentioned above very.
Although the single stage detector is faster than the two stage one, the location accuracy of the two stage detector
is marginally higher. However, there is a significant issue for practical application computationally expensive and
memory consuming. This problem exists independent of the type of detecting network [13].
In the field of goal recognition, YOLO algorithm has carried out well. In this paper, it enhance the brand new
YOLO community version via way of means of converting the residual devices to dense connection in the CSP
module and including channel interest mechanism. The advanced community version alleviates the vanishing-
gradient hassle, complements function propagation, encourages function reuse, and decreases the quantity of
parameters. What’s more, it can adaptively recalibrate the channel records of the function maps and enhance the
overall performance of goal detection. Experimental consequences display that the advanced YOLO community
version greatly improves the detection accuracy. In addition, it optimizes the hassle of lacking and miss-detecting
targets convolutional neural network (CNN) has end up the famous deep gaining knowledge of set of rules for
goal recognition. However, while the function maps attain the stop of the community structure after a couple of
convolutions, the function facts in it may disappear. To clear up this type of problem, pupils placed forward a few
answers and fashions, consisting of ResNet and DenseNet . Each of those models has a key function that they
devise a short route among the front community layer and the again community layer. In the ResNet community
model, channels are brought up, but channels are linked within the DenseNet community model.
It guarantees that the facts go with the drift among layers within the community is maximized. And as it
does now no longer want to analyze redundant function maps, every layer community can attain extra function
facts with fewer parameters. In addition, DenseNet's facts go with the drift and gradients throughout the
community are improved, in order that it turns into less difficult to be trained [14].
Convolution neural networks [CNN] were widely used in photographs and other parts as artificial intelligence
(AI) advanced due to their superior performance. Although CNN faces significant challenges in the development
of mobile devices due to its computationally complex set of algorithms. FPGA are good candidates for CNN
deployment because of their high performance, re-programmability, and sporadic energy usage. The YOLO set
of rules views goal detection as a regression issue in comparison to various CNN algorithms. It is a one-step set
of rules that execute quickly and require little calculation it is suitable for hardware platform to implement. This
paper suggests a sophisticated set of guidelines for a deep learning YOLO community that are entirely based on
Xilinx ZYNQ FPGA. The issue of large computational of CNN and confined assets on FPGA chips as resolved
by improving the YOLO community version and fixed point, etc. The parallelism FPGA is then exploited to
increase the seat and length. Results from experiments show that the method suggested in this study as greatly
improved operational cost while maintaining accuracy and as a very reasonable cost inside the consideration of
CNN real time computing [15].
In this paper, the author proposed a method to classify human and automotive objects by integrating a support
vector machine (SVM) with the deep learning version, you only look once (Yolo), in a high resolution automotive
radar device. The constraints of gold envision from the yellow version or incorporated to the SVM in order to
improve the overall performance of the class. The result from the yolo and SVM may then be combined with the
predetermined target bonds to advance the overall class accuracy. The results showed that they suggested method
outperforms those obtained using yolo or in terms of overall performance.
A new device version that combines the conventional method, an SVM, and the deep learning version,
you only look once (yolo), to categorize people and motors. Based on real measurements, the proposed technique
can classify people and motors with over 90% accuracy in high decision automotive radar will stop using the
suggested version, the SVM and yolo version are training using range angle (RA) domain information from
preprocessed radar signal to classify the same scene. In particular, the enhanced class performance by utilizing
boundary containers from the yolo version. Finally, by integrating every result, the overall class accuracy is
increased [16].
In this paper the author describes the CPU based yolo, a real time object detection version that may be used with
non GPU computers to benefit users of low configuration computers. Many well developed algorithms are
available for object detection including yolo, Faster R-CNN, Fast R-CNN R, R-CNN, mask R-CNN, R-FCN,
SSD, Retinanet, and others. YOLO is a set of deep neural network rules for object detection that is more accurate
and quick than most other algorithms. YOLO is made of four computers that are entirely GPU based and media
graphics card with at least 12 GB of memory. In this paper they are implemented yolo with open CV in order to
make real time object detection possible on CPU based computers. On several non GPU computers, our version
accurately and reliably detects an object from video at a frame rate between 10.12 and 6.29 frame per second. The
mAP for CPU based Yolo is 31.5% the evaluation of computer vision using deep learning as being built up and
completed throughout time mostly using one specific, well known set of rules called convolution neural networks.
It has a convolution layer where the output is produced by convolving a clear output with a variety of input
components. The environment of a convolution layer allows for the extraction of related styles from companion
Journal of Xi'an University of Architecture & Technology
Volume XIV, Issue 7, 2022
ISSN No : 1006-7930
Page No: 148
6. input as shown in fig.3. Furthermore, compared to a tightly linked layer, a convolution layer tends to require a
considerably less weight to be determined because filter are not used in convolution layer [17].
Figure.3 Convolutional Neural Networks design
This study builds a smart monitoring device in response to the current status of the monitoring device, which
requires human management and isn't intelligent enough. The device completes the evaluation of the pictures
content, implement smart monitoring, and uses deep learning to distinguish things. The program of the smart
monitoring gadget is realized using the twisting moment as an example. The yolo V2 version curling automated
snapping images device detection record set throughout the test had a precision rate of 94.3%, and the detection
speed was 13 frame per second. The yolo version can accommodate the curling the photo taking devices real time
requirements. That equipment completed its role of monitoring during the actual photographic process. For the
first time, it is applied in this paper to curling gold detection underneath the curling smart monitoring equipment
the digital camera platform is operated at the same time based on Yolo V2 curling detection result. The smart
monitoring of the curling is completed and combined with this straightforward recording of the curling track.
Study shows that yolo V2 fully satisfies the real time requirements with detection accuracy for curling goals of
94.3% and detection rate of 30 frames per second. The yolo based fully curling smart monitoring equipment can
guarantee accuracy and real time performance at the same time as well as complete the actual photographic task.
The content smart assessment model of the digital camera previews is introduced by a device architecture of the
curling smart monitoring device. The teaching of Yolo apply to the smart evaluation version is introduced in the
third element along with the evaluation of outcomes [18].
Real-time detection of the pavement surroundings is an essential if part of independent riding technology. This
paper gives a real time automobile detection device primarily based totally on embedded devices. Based on the
prevailing YOLOv3-tiny neural community shape, this paper proposes a brand new neural community shape the
YOLOv3- tiny, and quantify the community parameter in the community. Trading complexity of competing in
embedded devices, making the proposal neural community shape more appropriate for embedded devices. The
new shape is tested, earlier than quantization the YOLO V3 live’s detection precision way of means of the
convolution operation and dispatched to the Yolo layer for goal detection. At the same time, the 13x13 function
map is merged with the 26x26 function map is dispatched second one yolo layer for goal detection. The entire
network shape makes use of yolo layers to hit upon exceptional size of function graphics, in order that the
community version can hit upon last targets in addition to small targets. The community shape of YOLO V3 tiny
can gain 87.79% mAp at the take a look at set, after quantization of community parameters can gain 69.79% mAP
and detection pays can gain 28 FPS. The layer shape of Yolo V3 that is specifically composed of convolution
layer and pool layer as shown in fig.2. A small function map of 13x13 is received via way of means of the usage
of pooling layer to carry out 5 down sampling operation at the input picture, after which the function is extracted
[19].
Figure.4 The structure of YOLOv3-tiny network
Journal of Xi'an University of Architecture & Technology
Volume XIV, Issue 7, 2022
ISSN No : 1006-7930
Page No: 149
7. Breed type is type of detected animal with the aid of using the usage of the pc imaginative and prescient method.
This observe is accomplished to assess the overall performance of livestock detection and breed type consistent
with particular areas at the frame of livestock that differ consistent with their breed. You Only Look Once (YOLO)
the set of rules is used for livestock detection and breed type on the information set. A unique information set is
generated using images gathered from Google images in order to demonstrate the usefulness of the suggested
strategy in the detection and type of livestock. Experimental results show that the YOLO set of rules can accurately
identify the breed of animals on autograph with 92.85% accuracy. With the use of item detection set of rules
YOLOv4, this observation seeks to livestock dictation and type. The item detection set of criteria, however,
provides the detection and type of the cattle in the photograph. YOLO are used to identify livestock and categories
breed for this purpose. The data set was manually constructed from images collected from various structure
because there may not be a formally educated version and information sector version education, the title said was
manually generated using photographs collected from various structures. As it expand this data set, it pay
particular attention to the gender of animals like calf, cow, and steer. The data set was manually classified after
that. The positioning and sophistication of the creatures inside the image are determined as a result of being this
tagging. Later, version education is accomplished with the aid of using feeding the version with the pc imaginative
and prescient method used in this observe. The experimental outcomes display that the educated version finished
92.85 accuracy at the given take a look at set [20]-[23].
III. CONCLUSION
In addition to providing an overview of the real time object detection technique used by YOLO, this paper
discusses the core CNN algorithm structure. CNN architecture models have the ability to eliminate highlights and
identify objects any given image. When implemented appropriately, CNN models can address issues like
deformity diagnosis, creating educational or instructive application, etc. In practice, yolo has a lot of advantages
over other CNN algorithm. Yolo can train the complete model in parallel since it is a unified object detection
model that is easy to build and train in accordance with its simple loss function. The best speed and accuracy
tradeoff for object detection is offered by second major version of YOLO known as YOLOv4. In addition, Yolo
is the most advanced object identification technique and is advised for real time object detection since it
generalized object representation more effectively than other object detection models. With these
accomplishments, it is clear that field of object detection as a bright future.
REFERENCES
1. Peiyuan Jiang, Daji Ergu*, Fangyao Liu, Ying Cai, Bo Ma ; A Review of Yolo Algorithm Developments ;The 8th International
Conference on Information Technology and Quantitative Management ( ITQM 2020 & 2021)
2. Joseph Redmon∗ , Santosh Divvala∗, Ross Girshick , Ali Farhadi; “You Only Look Once: Unified, Real-Time Object Detection” ;
University of Washington∗ , Allen Institute for AI† , Facebook AI Research,9 may 2016
3. Geethapriya. S, N. Duraimurugan, S.P. Chokkalingam; “Real-Time Object Detection with Yolo” International Journal of Engineering and
Advanced Technology (IJEAT) ISSN: 2249 – 8958, Volume-8, Issue-3S, February 2019.
4. Tanvir Ahmad , 1 Yinglong Ma , 1 Muhammad Yahya,2 Belal Ahmad,3 Shah Nazir , 4 and Amin ul Haq; “Object Detection through
Modified YOLO Neural Network” Research Article, 6 June 2020.
5. Zhenguang Li; “An improved algorithm for deep learning YOLO network based on Xilinx ZYNQ FPGA”; 2020 International Conference
on Culture-oriented Science & Technology (ICCST).
6. JIN Zhao-zhao1, ZHENG Yu-fu1; “Research on Application of Improved YOLO V3 Algorithm in Road Target Detection”; ICMTIM
2020 Journal of Physics: Conference Series.
7. Shuo Wang; “Research Towards Yolo-Series Algorithms: Comparison and Analysis of Object Detection Models for Real-Time UAV
Applications”; IoTAIMA 2021 Journal of Physics: Conference Series.
8. Priya Kumari , Sonali Mitra , Suparna Biswas, Sunipa Roy, Sayan Roy Chaudhuri, Antara Ghosal, Palasri Dhar, Anurima Majumder;
“YOLO Algorithm Based Real-Time Object Detection”; June 2021| IJIRT | Volume 8 Issue 1 | ISSN: 2349-6002
9. Dr. N. Murali Krishna1 , Ramidi Yashwanth Reddy2 , Mallu Sai Chandra Reddy3 , Kasibhatla Phani Madhav4 , Gaikwad Sudham;
“OBJECT DETECTION AND TRACKING USING YOLO ”; 2021 Third International Conference on Inventive Research in Computing
Applications (ICIRCA) .
10. Rakha Asyrofi, Yoni Azhar Winata ; “The Improvement Impact Performance of Face Detection Using YOLO Algorithm”; Proc. EECSI
2020 - 1-2 October 2020
11. Rekha B. S.1, Athiya Marium2, Dr. G. N. Srinivasan3, Supreetha A. Shetty; “Literature Survey on Object Detection using YOLO”;
International Research Journal of Engineering and Technology (IRJET) , Volume: 07 Issue: 06 | June 2020.
12. Zhong-Qiu Zhao, Member, IEEE, Peng Zheng, Shou-tao Xu, and Xindong Wu, Fellow, IEEE; “YOLOv3: An Incremental Improv Object
Detection with Deep Learning: A Review”; 16 Apr 2019
13. Yonghui Lu1, Langwen Zhang2, Wei Xie; “YOLO-compact: An Efficient YOLO Network for Single Category Real-time Object
Detection”; 2020 IEEE
Journal of Xi'an University of Architecture & Technology
Volume XIV, Issue 7, 2022
ISSN No : 1006-7930
Page No: 150
8. 14. Fengxi Yan, Yinxia Xu; “Improved Target Detection Algorithm Based on YOLO”; 2021 4th International Conference on Robotics,
Control and Automation Engineering
15. Zhenguang Li; “An improved algorithm for deep learning YOLO network based on Xilinx ZYNQ FPGA”; 2020 International
Conference on Culture-oriented Science & Technology (ICCST).
16. Woosuk Kim , Hyunwoong Cho , Jongseok Kim , Byungkwan Kim , and Seongwook Lee; “Target Classification Using Combined YOLO-
SVM in High-Resolution Automotive FMCW Radar”; 2020 IEEE Radar Conference.
17. Md. Bahar Ullah; “CPU Based YOLO: A Real Time Object Detection Algorithm”; 2020 IEEE Region 10 Symposium (TENSYMP), 5-7
June 2020, Dhaka, Bangladesh.
18. Shujun Zhang, Shanzhen Lan, Qi Bu, Shaobin Li; “YOLO based Intelligent Tracking System for Curling Sport”; ICIS 2019, June 17-19,
2019
19. Shaobin Chen1 ,Wei Lin11; “Embedded System Real-Time Vehicle Detectionbased on Improved YOLO Network”;2019 IEEE 3rd
Advanced Information Management,Communicates,Electronic and Automation Control Conference (IMCEC 2019)
20. Gizem Nur Uzun, Alperen Yılmaz, Mustafa Zahid Gürbüz, Oğuzhan Kıvrak; “Detection and Breed Classification of Cattle Using YOLO
v4 Algorithm”; 2021 International Conference on INnovations in Intelligent SysTems and Applications (INISTA)
21. Viswanatha, V., et al. "Intelligent line follower robot using MSP430G2ET for industrial applications." Helix-The Scientific Explorer| Peer
Reviewed Bimonthly International Journal 10.02 (2020): 232-237.
22. Viswanatha, V., Ashwini Kumari, and Pradeep Kumar. "Internet of things (IoT) based multilevel drunken driving detection and prevention
system using Raspberry Pi 3." International Journal of Internet of Things and Web Services 6 (2021).
23. Viswanatha, V., A. C. Ramachandra, and R. Venkata Siva Reddy. "Bidirectional DC-DC converter circuits and smart control algorithms:
a review." Journal of Electrical Systems and Information Technology 9.1 (2022): 1-29.
Journal of Xi'an University of Architecture & Technology
Volume XIV, Issue 7, 2022
ISSN No : 1006-7930
Page No: 151