Method, device and system for testing target recognition algorithm
A target recognition and testing method technology, applied in the field of image recognition, to achieve accurate and reliable evaluation
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0031]First, refer to figure 1 An example electronic device 100 for implementing the testing method, device and system of the target recognition algorithm of the embodiment of the present invention will be described.
[0032] Such as figure 1 As shown, the electronic device 100 includes one or more processors 102, one or more storage devices 104, an input device 106, an output device 108, and an image acquisition device 110, and these components are connected through a bus system 112 and / or other forms of connection mechanisms (not shown) interconnects. It should be noted that figure 1 The components and structure of the electronic device 100 shown are only exemplary, not limiting, and the electronic device may also have other components and structures as required.
[0033] The processor 102 may be a central processing unit (CPU) or other forms of processing units with data processing capabilities and / or instruction execution capabilities, and may control other components i...
Embodiment 2
[0040] refer to figure 2 A flow chart of a test method for a target recognition algorithm shown, the method may include the following steps:
[0041] Step S202, acquiring frame images of the video to be tested, where the video to be tested contains the target object.
[0042] In an embodiment of the present invention, an implementation manner of acquiring a frame image of a video to be tested is given: (1) Acquire a video to be tested that includes a target object. In practical applications, the video to be tested can be collected by a camera device. (2) Decode the video to be tested to obtain each frame of image. For example, the FFmpeg tool can be used to decode the video frame by frame, and the picture of each frame in the video stream is cut and saved in units of preset periods. The preset cycle can be 25 frames per second, and of course other values can also be set. (3) Arranging frame images in time sequence. Each frame of images can be arranged with frame number...
Embodiment approach
[0046] (1) Determine the first frame number of the frame image when the target object appears for the first time in the video to be tested, and the second frame number of the frame image of the target object first recognizable in the video to be tested.
[0047] (2) The first frame number and the second frame number are used as frame position information corresponding to the target object.
[0048] In practical applications, the benchmark identification parameters can also be recorded in the form of sequences. For example, target object - first frame number - second frame number.
[0049] Step S206, applying the target recognition algorithm to identify the target object in the video to be tested, and obtaining algorithmic recognition parameters of the target object. The algorithm identification parameters include the frame position information recorded during the identification process of the target identification algorithm.
[0050] The following is an implementation manner...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More