This is a presentation that implements functions related to autonomous driving in driving video.
Open cv and tensorflow object detection api.
lane detection was performed using opencv,
Vehicle detection was performed using a pre-trained model,
traffic light detection used transfer learning.
12. 12
02
Feature
1 Lane Detection
Undistortion
- Compute the camera calibration matrix and distortion coefficients
given a set of chessboard images
- Apply a distortion correction to raw images
15. 15
02
Feature
1 Lane Detection
Measuring & Projecting
- Determine curvature of the lane (Use Radius of Curvature)
- Warping the detected lane boundaries back onto the original
image
- Output visual display of the lane boundaries
16. 16
02
Feature
2 Vehicle Detection
Using Tensorflow Object detection API Model zoo
Model name Speed
( ms )
Acc
(mAP)
Model
size
Ssd_mobilenet_v2 31 22 201m
Ssd_inception_v2 42 24 295m
Faster_rcnn_inception_v2 58 28 167m
Faster_rcnn_resnet50 89 30 405m
Faster_rcnn_resnet101 106 32 624m
Rfcn_resnet101 92 30 685m
17. 17
02
Feature
2 Vehicle Detection
Coco dataset Instance
person
bicycle
car
motorcycle
airplane
bus
train
truck
boat
trafficlight
firehydrant
stopsign
parkingmeter
bench
bird
cat
dog
horse
sheep
cow
elephant
bear
zebra
giraffe
backpack
umbrella
handbag
tie
suitcase
frisbee
skis
snowboard
sportsball
kite
baseballbat
baseballglove
skateboard
surfboard
tennisracket
bottle
wineglass
cup
fork
knife
spoon
bowl
banana
apple
sandwich
orange
broccoli
carrot
hotdog
pizza
donut
cake
chair
couch
pottedplant
bed
diningtable
toilet
tv
laptop
mouse
remote
keyboard
cellphone
microwave
oven
toaster
sink
refrigerator
book
clock
vase
scissors
teddybear
hairdrier
toothbrush
person
bicycle
car
motorcycle
airplane
bus
train
truck
boat
trafficlight
firehydrant
stopsign
parkingmeter
bench
bird
cat
dog