Unlock instant, AI-driven research and patent intelligence for your innovation.

Model test method and device, electronic equipment and storage medium

A test method and model technology, applied in the computer field, can solve the problems of low test efficiency and accuracy, high manual workload, etc., and achieve the effect of solving more manual workload and improving test efficiency and accuracy

Pending Publication Date: 2021-11-12
SHENZHEN HONGDIAN TECH CORP
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the traditional testing methods for target recognition models have technical problems such as high manual workload, low testing efficiency and low accuracy.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Model test method and device, electronic equipment and storage medium
  • Model test method and device, electronic equipment and storage medium
  • Model test method and device, electronic equipment and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0029] figure 1 It is a schematic flow chart of a model testing method provided in Embodiment 1 of the present invention. This embodiment is applicable to testing the pre-trained target recognition model to determine whether the target recognition model to be tested is the expected model In some cases, the method can be executed by a model testing device, which can be realized by software and / or hardware, and the model testing device can be integrated into electronic equipment such as a computer or a server.

[0030] Such as figure 1 As shown, the method of the present embodiment includes:

[0031] S110. Acquire the video to be tested, determine the analyzed video frame corresponding to each scene in the video to be tested, perform feature extraction on the analyzed video frame corresponding to each scene, and obtain a corresponding expected feature image.

[0032] Wherein, the video to be tested may be a video to be tested at the current moment. The number of videos to be ...

Embodiment 2

[0073] image 3 It is a schematic flow chart of a test method for a model provided by Embodiment 2 of the present invention. On the basis of the foregoing embodiments, optionally, the expected feature images and actual output features corresponding to the analyzed video frames included in each scene image, obtaining the scene recognition rate of the video to be tested, comprising: obtaining the scene recognition rate of each scene according to the expected feature image corresponding to the analysis video frame contained in each scene and the actual output feature image; The scene recognition rate determines the scene recognition rate of the video to be tested.

[0074] Wherein, technical terms that are the same as or corresponding to those in the foregoing embodiments will not be repeated here.

[0075] Such as image 3 As shown, the method in this embodiment may specifically include:

[0076] S210. Acquire the video to be tested, determine the analyzed video frame corresp...

Embodiment 3

[0109] Figure 4 It is a schematic diagram of a model test device module provided in Embodiment 3 of the present invention. The present invention provides a model test device, which includes: an expected feature image acquisition module 310, an actual output feature image acquisition module 320, and a scene recognition rate Obtain the module 330 and the target recognition model determination module 340 to be tested.

[0110] Wherein, the expected feature image obtaining module 310 is used to obtain the video to be tested, determine the analysis video frame corresponding to each scene in the video to be tested, perform feature extraction on the analysis video frame corresponding to each scene, and obtain the corresponding expected feature Image: The actual output feature image obtaining module 320 is used to input the analysis video frame corresponding to each scene into the pre-trained target recognition model to be tested, and obtain the actual output feature corresponding to...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The embodiment of the invention discloses a model testing method and device, electronic equipment and a storage medium. The method comprises the steps of obtaining a to-be-tested video, determining an analysis video frame corresponding to each scene in the to-be-tested video, and performing feature extraction on the analysis video frame corresponding to each scene to obtain a corresponding expected feature image; inputting the analyzed video frame corresponding to each scene into a pre-trained to-be-tested target recognition model to obtain an actual output feature image corresponding to each analyzed video frame contained in each scene; obtaining the scene recognition rate of the to-be-tested video according to the expected feature image and the actual output feature image corresponding to the analyzed video frame contained in each scene; according to the scene recognition rate of the to-be-tested video, whether the to-be-tested target recognition model is the expected model or not is determined, so that whether the target recognition model is qualified or not is determined more quickly, and the test efficiency and accuracy are improved.

Description

technical field [0001] The embodiments of the present invention relate to the field of computer technology, and in particular to a model testing method, device, electronic equipment and storage medium. Background technique [0002] Artificial intelligence is now used more and more maturely, such as face recognition, image recognition, etc. Image recognition has been applied to all aspects of our lives. Object recognition software installed on mobile phones can help us identify the names of various objects. Mobile payment applications Face recognition greatly improves the convenience and security of payment. The realization of more image recognition technologies requires more reliable and accurate testing methods. [0003] The traditional test method for the target recognition model needs to manually mark the target recognition object in each video frame in each scene of the video in advance, and then use the target recognition model to determine the target recognition objec...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/00G06K9/62G06N3/04
CPCG06N3/045G06F18/2193
Inventor 李仟
Owner SHENZHEN HONGDIAN TECH CORP