Vehicle retrieval method and apparatus
A vehicle and target vehicle technology, applied in the field of image processing, can solve the problem of low vehicle retrieval accuracy, and achieve the effects of high vehicle retrieval accuracy, low missed detection rate, and high accuracy
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0059] This embodiment provides a vehicle retrieval method, which is used in a vehicle retrieval device. like figure 1 As shown, the vehicle retrieval method includes the following steps:
[0060] In step S11, an image of the vehicle to be detected is acquired.
[0061] The image of the vehicle to be detected in this embodiment may be stored in the vehicle retrieval device in advance, or the vehicle image acquired by the vehicle retrieval device from the outside in real time, or the vehicle image extracted from an image by the vehicle retrieval device.
[0062] Step S12, extracting N detection feature vectors of the vehicle to be detected from the image of the vehicle to be detected, where N is greater than or equal to 2, and each detection feature vector corresponds to a component of the vehicle to be detected.
[0063] In this example, the vehicle retrieval device extracts N detection feature vectors of the vehicle to be detected from the image of the vehicle to be detecte...
Embodiment 2
[0083] This embodiment provides a vehicle retrieval method, which is used in a vehicle retrieval device. like figure 2 As shown, the vehicle retrieval method includes the following steps:
[0084] Step S20, acquiring an image of the vehicle to be detected.
[0085] In this embodiment, after acquiring the image of the vehicle to be detected, the image of the vehicle to be detected is format-converted, and the image in the first format is converted into an image in the second format, which facilitates subsequent calculation processing and improves the efficiency of vehicle retrieval. For example, if the acquired image of the vehicle to be detected is in RGB format, it is converted into YUV format.
[0086] The rest are the same as step S11 in Embodiment 1, and are not repeated here.
[0087] Step S21, extracting N detection feature vectors of the vehicle to be detected from the image of the vehicle to be detected, where N is greater than or equal to 2, and each detection fea...
Embodiment 3
[0116] This embodiment provides a vehicle retrieval method, which is used in a vehicle retrieval device. like Figure 4 As shown, the vehicle retrieval method includes the following steps:
[0117] Step S311, acquiring a plurality of images including vehicles.
[0118] In this embodiment, the vehicle retrieval device acquires a plurality of images including vehicles from the input electric police or bayonet images.
[0119] Step S312, extracting vehicle images from a plurality of images containing vehicles by using a convolutional neural network model.
[0120] Among them, the CNN model used to extract vehicle images is one of RCNN, Fast RCNN, Faster RCNN, YOLO, and SSD.
[0121] Step S313 , extracting vehicle features from the extracted vehicle image by using a convolutional neural network model, and establishing corresponding relationships for different features of the same vehicle to obtain a feature vector library.
[0122] Among them, the CNN model used to extract veh...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More