Image ROI positioning method and device based on dimension invariant feature transformation during automobile instrument panel visual detection

A scale-invariant feature, automotive dashboard technology, applied in image analysis, image data processing, instruments, etc., can solve the problem of the influence of image features to be detected, poor ROI positioning effect, etc., to avoid the influence of ROI positioning results Good stickiness and ROI positioning effect, avoiding different effects of feature point sets

Inactive Publication Date: 2014-10-08
HARBIN INST OF TECH
View PDF2 Cites 11 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] The purpose of the present invention is to solve the problem that the features of the image to be detected will be affected to a large extent when the current template-based ROI positioning method is dealing with too bright, too dark and blur

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
  • Image ROI positioning method and device based on dimension invariant feature transformation during automobile instrument panel visual detection
  • Image ROI positioning method and device based on dimension invariant feature transformation during automobile instrument panel visual detection
  • Image ROI positioning method and device based on dimension invariant feature transformation during automobile instrument panel visual detection

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0036] Specific implementation mode one: combine figure 1 Describe this embodiment, the image ROI positioning method based on scale-invariant feature transformation in the automobile dashboard visual detection described in this embodiment, it comprises the following steps:

[0037] A step for obtaining the ROI and positioning ROI information of the vehicle dashboard template image and the detection item area of ​​the image to be detected, and storing the information;

[0038] A step for extracting a set of feature points and a set of feature vectors of the template image using a scale-invariant feature transformation method according to the ROI for positioning of the template image, and saving them;

[0039]A step for extracting a set of feature points and a set of feature vectors of the image to be detected by using a scale-invariant feature transformation method according to the ROI for positioning of the image to be detected, and saving;

[0040] For finding the n-dimensio...

specific Embodiment approach 2

[0057] Specific embodiment 2: This embodiment is a further limitation of the image ROI positioning method based on scale invariant feature transformation in the visual detection of automobile instrument panel described in specific embodiment 1. Item and the n-dimensional space Euclidean distance of each item of the feature vector set of the image to be detected, and the steps of extracting the initial matching point pair set by the Euclidean distance include:

[0058] It is used to select a feature vector from the feature vector set of the template image, and take the Euclidean distance from all feature vectors in the feature vector set of the image to be detected, where the minimum distance is Edis 1 , the next smallest distance is Edis 2 , when Edis 1 ≤β*Edis 2 , then the corresponding two feature points are the steps of the initial matching point pair

[0059] The β is the Euclidean distance multiple threshold value, and the value range of β is 0.5-0.9;

[0060] It is u...

specific Embodiment approach 3

[0066] Embodiment 3: This embodiment is a further limitation of the image ROI positioning method based on scale-invariant feature transformation in the visual detection of automobile dashboard described in Embodiment 2, β=0.75.

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 provides an image ROI positioning method and device based on dimension invariant feature transformation during automobile instrument panel visual detection, belongs to the field of image ROI positioning and aims at solving the problem that at present, an ROI positioning method based on a template is poor in effect. The image ROI positioning method comprises the steps of confirming ROI of detection item regions of a template image and an image to be detected, extracting feature point sets and feature vector sets of the template image and the image to be detected by adopting the image ROI positioning method based on the dimension invariant feature transformation, utilizing an Euclidean distance to perform initial matching on the template image and the image to be detected, using a Mahalanobis distance to screen matching point pairs, further obtaining transformation matrixes of matching points from the template image to the image to be detected, performing reconstitution on the ROI of the detection item regions of the image to be detected according to the transformation matrixes to obtain newest ROI and detecting the newest ROI to judge the positioning success situation. The image ROI positioning method and device is used for automobile instrument panel visual detection.

Description

technical field [0001] The invention belongs to the field of image ROI positioning. Background technique [0002] In the visual detection of automobile dashboard, the first step is to select the ROI (Region of Interest) of the image. There are many items to be detected on the car dashboard, and different regions have different ROIs. These ROIs need to accurately contain the region to be detected and cannot overlap with each other. In the assembly line work, the position of the camera often has different degrees of deviation, and these deviations are difficult to completely eliminate with some mechanical facilities. These position deviations will make the original ROI of the image unable to select the correct region to be detected, which will cause unpredictable errors in subsequent image processing. [0003] Software ROI positioning based on image processing can well eliminate the adverse effects of position deviation on subsequent processing. Compared with manually placi...

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): G06T7/00
Inventor 高会军李博伦于金泳李竹奇杨学博由嘉
Owner HARBIN INST OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap