Measurement of vehicle distance of its front-axle vision inspection

Scene: A camera in front of the car, with visual information to detect lane lines and vehicles in front to assist in driving

Lane line detection: usually straight lines or curves, white, using grayscale or grayscale changes. First seek the edge, then Hough detect straight line or use. Depending on the scene, morphological operations may be added to remove noise;

Front detection: Often the picture is the rear of the car, the scene may be day or night.
Daytime can be divided into pattern recognition (statistical) method and the method of detecting the contour of the target; statistical ideas, in order to live, with Viola and Jones Cascade classifier. Direct detection of the target, that is, the detection of the car's tail, often detect the shadow between the car and the road, because this piece of gray is usually very low, and the scene in other areas of the Cvblob, the method of large-scale. Symmetry can also be exploited.
At night, the main use of lights, symmetry, luminance value of information

Can be traced after detection. TLD or KF?

License plate location: Commonly used license plate for the blue bottom white, yellow black word, mainly in the area gray changes are obvious, so you can detect the edge, the edge of the dense place for the license plate possible area. Vertical projection method to find the left and right edges

Distance measurement: Assuming that the camera's height and pitch angle are known, using the point in front of the bottom of the vehicle, in the world coordinate system is located on the ground (assuming ground level), you can find out the point and camera distances

Measurement of vehicle distance of its front-axle vision inspection