Image identification method
A technology of image recognition and recognition method, which is applied in the field of image recognition, can solve problems such as collision of vehicles in front, lack of target type, and inability to accurately determine vehicles in front, and achieve the effect of little noise influence
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0026] A kind of image recognition method, its recognition method comprises the following steps:
[0027] A. Obtain the driving image of the vehicle to be identified;
[0028] B. Identify and record the overall shape of the vehicle and the bottom edge of the rear car in the image of the vehicle to be identified, and enter the information into the in-vehicle analysis sensor and computing center;
[0029] C. Calculate the distance between the analyzed and processed vehicle information and the driving vehicle, and obtain accurate data.
Embodiment 2
[0031] In embodiment one, add following operation:
[0032] In Step A, the following information will be identified in detail:
[0033] a. Vehicle geometry: When a road vehicle is viewed from the rear, it contains many horizontal structures such as rear windows, trunks, and bumpers. A vehicle detection method based on geometric structures can also be proposed; daytime vehicle positioning needs to consider the shadow problem under the vehicle. Its advantage is that all potential vehicles can be detected;
[0034] b. Shadows under the vehicle: the texture of the vehicle is smoother than that of the surrounding environment; almost all wheels are along the central axis, this feature can well distinguish the vehicle from the environment, and is less affected by noise;
[0035] c. Texture symmetry: almost all wheels are along the central axis, this feature can well distinguish the vehicle from the environment, and is less affected by noise;
[0036] d. Color information taillights...
Embodiment 3
[0042] In embodiment two, add following operation:
[0043] In step B, all the information of the vehicle entered into the computing center is analyzed and processed, and before the target object in the driving image of the vehicle to be recognized is recognized, the driving image of the vehicle to be recognized is preprocessed.
[0044] A kind of image recognition method, its recognition method comprises the following steps:
[0045] A. Obtain the driving image of the vehicle to be identified;
[0046] B. Identify and record the overall shape of the vehicle and the bottom edge of the rear car in the image of the vehicle to be identified, and enter the information into the in-vehicle analysis sensor and computing center;
[0047] C. Calculate the distance between the analyzed and processed vehicle information and the driving vehicle, and obtain accurate data.
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 
