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

Method and device for car surface damage classification based on deep learning

A technology of automobile surface and classification method, which is applied in the field of automobile surface damage classification, and can solve the problems of complex and changeable appearance scratches.

Active Publication Date: 2018-10-16
高前文
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The exterior scratches of the car are complex and changeable, and are easily affected by external interference such as light and occlusion

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
  • Method and device for car surface damage classification based on deep learning
  • Method and device for car surface damage classification based on deep learning
  • Method and device for car surface damage classification based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0086] All features disclosed in this specification, or steps in all methods or processes disclosed, may be combined in any manner, except for mutually exclusive features and / or steps.

[0087] Any feature disclosed in this specification, unless specifically stated, can be replaced by other alternative features that are equivalent or have similar purposes. That is, unless expressly stated otherwise, each feature is one example only of a series of equivalent or similar features.

[0088] Deep learning is a machine learning theory that discovers distributed feature representations of data by combining low-level features to form more abstract high-level representation attribute categories or features. It can be divided into supervised learning and unsupervised learning. Convolutional neural network is a deep learning model under supervised learning. It is a non-fully connected neural network structure that can automatically learn target features containing a large amount of data....

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 invention relates to the field of image detection, in particular to a method and device for classifying automobile surface damage based on deep learning. Aiming at the problems existing in the prior art, the present invention provides a classification method and device. Carry out feature learning and classification on the input image to be tested, specifically using a region selective search algorithm to extract a candidate area from each image to be tested and record the position information of each candidate area; input the image to be tested and remove the output The feature map of the layer is extracted from the network model to extract the feature vectors of each candidate area of ​​the image to be tested; the feature vectors of each candidate area are input into the SVM classifier to find the target feature vector; according to the position of the target feature vector in the feature map, the The position of the corresponding candidate area on the image to be tested is the target area of ​​the image to be tested; the target area of ​​the image to be tested is input into the optimal classification network model, and the probability of the area at each damage level is output.

Description

technical field [0001] The invention relates to the field of image detection, in particular to a method and device for classifying automobile surface damage based on deep learning. Background technique [0002] In recent years, with the continuous development of urbanization in our country, the per capita car ownership in our country has continued to increase. According to the survey, as of 2015, the total number of cars in the country has exceeded 170 million. At the same time, the safety problems caused by cars are also increasing. After an accident, whether it is between motor vehicles or a collision between a vehicle and a fixed object, it will leave marks on the vehicle. These traces will seriously affect the appearance and use of the car, and the maintenance costs required for different types of traces are also different, so these traces need to be evaluated. At present, the evaluation of scratches on the exterior of a car mainly relies on the subjective judgment of ...

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 Patents(China)
IPC IPC(8): G06T7/00G06K9/62
CPCG06T7/0006G06T2207/30248G06T2207/20081G06F18/2411
Inventor 史方樊强王标
Owner 高前文
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