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

Target detection method and device

A target detection and target technology, applied in the direction of instruments, character and pattern recognition, computer parts, etc., can solve the problem of single label information and cannot be popularized and applied, and achieve the effect of reducing calculation amount, improving efficiency and promoting application

Active Publication Date: 2019-07-12
深兰人工智能应用研究院(山东)有限公司
View PDF12 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In this case, the labeling information of the target is relatively single, which cannot be well promoted and applied.

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
  • Target detection method and device
  • Target detection method and device
  • Target detection method and device

Examples

Experimental program
Comparison scheme
Effect test

example 1

[0100] determining a target distance map of the original image data according to the disparity feature map;

[0101] A target distance of the target is determined from the target distance map based on the coordinates of the target.

example 2

[0103] determining the disparity of the target from the disparity feature map according to the coordinates of the target;

[0104] According to the parallax of the target, the target distance of the target is determined.

[0105] Compared with Example 1, Example 2 can avoid calculating the target distance of each pixel data in the original image data, thereby reducing computational complexity and improving computational efficiency. The following uses Example 2 as an example to describe the calculation process of the target distance in detail. .

[0106] The determination of the coordinates of the target can be performed through the following steps, specifically:

[0107] Traverse the target candidate frame, and use the center point of the target candidate frame whose width and height are not zero as the target center point;

[0108] The disparity of the target is determined from the disparity feature map according to the coordinates of the center point of the target.

[010...

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 a target detection method and device, which are used for providing rich target labeling information. The method comprises the steps of obtaining original image data, and a target feature map and a parallax feature map of the original image data; according to the target feature map and the parallax feature map, determining a target feature and a target distance of a target included in the original image data; and fusing the target feature, the target distance and the original image data to obtain fused image data.

Description

technical field [0001] The present application relates to the field of artificial intelligence, in particular to a target detection method and device. Background technique [0002] With the continuous maturity of deep learning technology, Convolutional Neural Networks (CNN) is widely used in target detection and recognition, speech recognition, natural language processing and other fields, because it is in the field of target detection and recognition. [0003] At present, the existing target detection and target ranging are two sets of independent systems, that is, to realize target detection and target ranging, two sets of independent hardware support are required. In this case, the labeling information of the target is relatively single, which cannot be well promoted and applied. Contents of the invention [0004] An embodiment of the present application provides an object detection method for providing rich object labeling information. [0005] In the first aspect, t...

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): G06K9/32G06K9/62
CPCG06V10/255G06V2201/07G06F18/251
Inventor 陈海波
Owner 深兰人工智能应用研究院(山东)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