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

Target detection method and device, and computer storage medium

A target detection and target technology, applied in computing, image analysis, image enhancement, etc., can solve the problem that 2D target detection cannot provide all the information of the surrounding environment, and achieve the effect of improving accuracy

Active Publication Date: 2021-08-03
SHANGHAI GOLDWAY INTELLIGENT TRANSPORTATION SYST CO LTD
View PDF7 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0002] At present, 2D (2-dimension, two-dimensional) target detection such as traditional image recognition can only detect the specific category of the target and the position of the target in the current field of view, which has led to 2D target detection in the field of artificial intelligence such as intelligent transportation. It cannot provide all the information needed to perceive the surrounding environment, which makes 3D object detection more and more people's attention

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, and computer storage medium
  • Target detection method and device, and computer storage medium
  • Target detection method and device, and computer storage medium

Examples

Experimental program
Comparison scheme
Effect test

example 1

[0159] Example 1: Based on the point cloud semantic heat map and image semantic heat map, the target detection device determines adaptive fusion features.

[0160]In Example 1, based on the point cloud data collected for the detection area, a point cloud semantic heat map is predicted, and the point cloud semantic heat map can indicate the approximate distribution of the targets to be detected. At the same time, based on the camera images collected for the detection area, an image semantic heat map is predicted, and the image semantic heat map can also indicate the approximate distribution of the target to be detected. Then, feature fusion is performed based on the two semantic heat maps to obtain adaptive fusion features, and then the target is detected based on the adaptive fusion features. That is to say, in Example 1, firstly, according to different data sources, a fuzzy prediction is made on the target to be detected to obtain the prior knowledge of the target to be detec...

example 2

[0162] Example 2: Based on the point cloud semantic heat map and image features, the target detection device determines adaptive fusion features.

[0163] In Example 2, based on the point cloud data collected for the detection area, a point cloud semantic heat map is predicted, and the point cloud semantic heat map can indicate the approximate distribution of the targets to be detected. At the same time, the camera feature of the camera image is determined according to the camera image collected for the detection area. Then, feature fusion is performed based on the point cloud semantic heat map and image features to obtain adaptive fusion features, and then the target is detected according to the adaptive fusion features. That is to say, in Example 2, firstly, according to the point cloud data, a fuzzy prediction is made on the target to be detected to obtain the prior knowledge of the target to be detected under the point cloud data, and then the camera image features are fus...

example 3

[0165] Example 3: Based on the image semantic heat map and point cloud features, the target detection device determines adaptive fusion features.

[0166] In Example 3, an image semantic heat map is predicted according to the camera image collected for the detection area, and the image semantic heat map can indicate the approximate distribution of the target to be detected. At the same time, the point cloud features are determined based on the point cloud data collected for the detection area. Then, feature fusion is performed based on the image semantic heat map and point cloud features to obtain adaptive fusion features, and then the target is detected according to the adaptive fusion features. That is to say, in Example 3, a fuzzy prediction of the target to be detected is first performed based on the camera image to obtain the prior knowledge of the target to be detected under the camera image, and then the point cloud features are fused to further detect the target. Sinc...

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 target detection method and device, and a computer storage medium, and belongs to the technical field of artificial intelligence. In the target detection method of the application, a point cloud semantic thermodynamic diagram is predicted according to the point cloud data collected for the detection area, and the point cloud semantic thermodynamic diagram can indicate the approximate distribution condition of the to-be-detected target. And feature fusion is carried out based on the point cloud semantic thermodynamic diagram and the point cloud features to obtain adaptive fusion features, and a target is detected according to the adaptive fusion features. Namely, in the method of the invention, firstly, fuzzy prediction is carried out on the to-be-detected target according to the point cloud data to obtain the priori knowledge of the to-be-detected target, and then the target is further detected based on the predicted priori knowledge in combination with the accurate point cloud features. In this way, fuzzy detection can be carried out on the target in the target detection process, then accurate detection is carried out, and therefore the accuracy of the detected target is improved.

Description

technical field [0001] The embodiments of the present application relate to the technical field of artificial intelligence, and in particular to a target detection method, device, and computer storage medium. Background technique [0002] At present, 2D (2-dimension, two-dimensional) target detection such as traditional image recognition can only detect the specific category of the target and the position of the target in the current field of view, which has led to 2D target detection in the field of artificial intelligence such as intelligent transportation. It cannot provide all the information needed to perceive the surrounding environment, which makes 3D object detection more and more people's attention. 3D target detection specifically refers to: not only detecting the category of the target and the position of the target in the current field of view, but also detecting the length, width, height and rotation angle of the target in three-dimensional space. At present, 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): G06T7/00G06T3/00G06T7/73G06N3/04
CPCG06T7/0002G06T7/73G06T2207/20221G06T2207/10032G06N3/045G06T3/067Y02A90/10
Inventor 张泽瀚罗兵华
Owner SHANGHAI GOLDWAY INTELLIGENT TRANSPORTATION SYST CO LTD