Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Metamorphic test method based on correctness of intelligent traffic model

A technology of intelligent transportation and metamorphosis testing, applied in neural learning methods, software testing/debugging, biological neural network models, etc., can solve problems such as testing, difficulty in adapting to the correctness testing of intelligent transportation models, and difficulty in implementing traditional software testing

Pending Publication Date: 2020-10-30
深圳慕智科技有限公司
View PDF3 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, in the intelligent transportation model, there is no expected value in advance, and the output is some kind of prediction, which is difficult to test by comparing or verifying the prediction with some kind of expected value unknown in advance
Therefore, traditional testing methods are difficult to adapt to the correctness testing of intelligent traffic models
[0004] The unknown "oracle" in the intelligent traffic model application scenario makes it difficult to implement traditional software testing

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
  • Metamorphic test method based on correctness of intelligent traffic model
  • Metamorphic test method based on correctness of intelligent traffic model
  • Metamorphic test method based on correctness of intelligent traffic model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0021] This patent implements the correctness test of the intelligent traffic model through the metamorphosis test, mainly using the automatic generation technology of test samples, and the specific key technologies involved include CycleGAN model, deep convolutional neural network (CNN), target detection technology, etc.

[0022] CycleGAN model

[0023] In the present invention, we use the CycleGAN model to learn and train common scene samples and samples that satisfy the equivalent transformation relationship, obtain a generator that meets the requirements, and generate new samples that satisfy the equivalent transformation relationship through the generator. CycleGAN is a masterpiece of image conversion, and the sample data can be converted without pairing. For example, transforming zebras into horses, transforming models into cartoon characters, etc. The characteristic of CycleGAN is to first convert the image from one domain to another through a cycle, and then convert i...

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

According to the invention, a metamorphic test technology is used to carry out correctness test on the intelligent traffic model; a metamorphic relation applied to the intelligent traffic model is designed, the correctness of a program function is judged from multiple aspects, the intelligent traffic model is tested by using the original use case and a derivative use case generated based on the metamorphic relation, whether the output of the original use case and the derivative use case meets the corresponding metamorphic relation or not is judged, and a test result is obtained. According to the method, the correctness of the intelligent traffic model can be reasonably measured and evaluated by adopting a new software testing method.

Description

technical field [0001] The invention belongs to the field of model testing, in particular to the testing of the correctness of intelligent traffic models. Based on the input and output results of the intelligent transportation model, the metamorphic relationship is established, and new test cases are generated by using the metamorphic relationship. By verifying whether the metamorphic relationship is maintained, it is determined whether the test is passed and a test report is finally formed. Background technique [0002] Deep learning technology is more and more widely used in computer systems. In the field of intelligent transportation, deep learning models are also playing an extremely important role. At the same time, the testing for quality assurance of its software is bound to receive increasing attention. Traditional testing methods usually compare whether the expected output of the program is consistent with the actual output to determine the test results. However, ...

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
IPC IPC(8): G06F11/36G06N3/04G06N3/08
CPCG06F11/3684G06F11/3688G06N3/08G06N3/045
Inventor 陈振宇常家鑫夏志龙
Owner 深圳慕智科技有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com