image processing

1. RTIPAPPLIED TO TRAFFIC
–
QUEUE DETECTION ALGORITHM
BY
A.N.SAISUNITHA
ROLL NO : 11012D4503

2. CONTENTS
1. Introduction
2. Image analysis System Structure
3. Queue detection algorithm
4. Motion detection algorithm
5. Vehicle detection algorithm
6. Advantages of SMED
7. Left limit Selection program
8. Threshold selection program
9. Traffic movements at junction
10. Conclusion

3. INTRODUCTION
 Digital processing is done with a digital computer or any
microcontroller based system.
 Aimed to measure queue parameters accurately.
 Queue detection algorithm has two operations :
• vehicle detection
• motion detection.

4. DIGITAL IMAGE PROCESSING:

5. Need for processing of traffic data:
Traffic surveillance and control, traffic management, road safety
and development of transport policy.
Traffic parameters measurable:
Traffic volumes, Speed, Headways, Inter-vehicle gaps, Vehicle
classification, Origin and destination of traffic, Junction turning.

6. IMAGE ANALYSIS SYSTEM STRUCTURE

7. Two jobs to perform :
When Green light is on:
Determine no. of vehicles moving along particular
lanes and their classification by shape and size.
When Red light is on:
Determine the backup length along with the
possibility to track its dynamics and classify vehicles in
backup.

8. Queue Detection Algorithm :
A spatial-domain technique is used to detect queue
- implemented in real-time using low-cost system.
Two different algorithms are used with the technique
• Motion detection operation
• Vehicle detection operation

10. Motion detection operation:
Differencing two consecutive frames.
Histogram of the key region parts of the frames is
analyzed by comparing with the threshold value.
Key region should be at least 3-pixel-wide profile of
the image along the road.

11. A median filtering operation is firstly applied to
the key region (profile) of each frame and one-
pixel-wide profile is extracted.
Difference of two profiles is compared to detect
for motion.
When there is motion, the differences of the
profiles are larger than the case when there is no
motion. The motion can be detected by selecting a
threshold value.

12. MOTION DETECTION ALGORITHM

13. VEHICLE DETECTION METHODS
 Two main approaches :
1. Background based methods - Perform the difference of the current frame
with an updated background image
2. Gradient based methods – It is based on the detection of vehicle edges.
Guarantees a very fast computation and good results
Sensitive to noise and are used with filtering operators
 Two types of edge detectors :
1. Morphological based edge detectors
2. Gradient based techniques

14. VEHICLE DETECTION ALGORITHM
 A newly developed morphological edge detection operator is
Separable Morphological Edge Detector (SMED).
 STEPS :
 Gray Scale Conversion
 Filtering
 Boundary extraction using thresholding
 Make image one pixel thick in horizontal and vertical direction
 Remove the noisy pixels from the image and filling the pixel break in regular
pattern

15. Edge detectors consisting of separable medium filtering and
morphological operators.
The SMED approach is applied (f) to each sub-profile of the
image and the histogram of each sub-profile is processed by
selecting Dynamic left-limit value and a threshold value to detect
vehicles.
SMED has lower computational requirement while having
comparable performance to other morphological operators
SMED can detect edges at different angles, while other
morphological operators are unable to detect all kinds of edges.
SMED :

16. Image captured by camera SMED operator applied to the
image

17. This program selects a grey value from the
histogram of the window, where there are approx.
zero edge points above this grey value.
When the window contains an object, the left-limit
of the histogram shifts towards the maximum grey
value, otherwise it shifts towards the origin.
This process is repeated for a large no. of
frames(100),and the minimum of the left-limit of
these frames is selected as the left-limit for the
next frame.
LEFT-LIMIT SELECTION PROGRAM

18. Dyanamic Threshold selection program:
The no. of edge points greater than the left limit grey value of each
window is extracted for a large no. of frames .
These nos. are used to create a histogram where its horizontal and
vertical axes correspond to the no. of edge points greater than left limit
and the frequency of repetition of these numbers for a period of operation
of the algorithm .
This histogram is smoothed using a median filter and we expect to get
two peaks in the resulted diagram, one peak related to the frames passing
a vehicle and the other related to the frames without vehicles for that
window.

19. TRAFFIC MOVEMENTS AT JUNCTIONS (TMJ)
 Measuring traffic movements of vehicles at junctions such as number
of vehicles turning in a different direction.
 STEPS :
 Cover the boundary of the junction by a polygon
 Define a minimum numbers of key regions inside the boundary.
 Application of the vehicle detection on each profile.
 A status vector is created for each window in each frame.
 If a vehicle is detected in a window, a “one” is inserted on its corresponding
status vector, otherwise, a “zero” is inserted.
 A group of One’s corresponds to No.of vehicles
 A group of Zero’s corresponds to distance between the vehicles.

20. CONCLUSION
 Algorithm measure basic queue parameters.
 The algorithm uses simple and effective operations.
 To reduce computation time motion detection operation is applied first.
 The vehicle detection operation is a less sensitive edge based
technique.
 Queue length measurement shows 95% accuracy at maximum.

21. Thank you

25. THEORY BEHIND
 The camera parameters (a11, a12,…a33) are calculated by knowing –
 Co-ordinates of any 6 reference points of the real-world condition.
 Co-ordinates or their corresponding images.
 This equation is used to reduce the sizes of the sub-profiles.
 The length of sub-profile should be about length of the vehicle.