Target detection method and device, and computer readable storage medium
A target detection and target technology, applied in computer parts, calculations, instruments, etc., can solve the problems of reducing detection accuracy, influence of detection results, false positives, etc., so as to reduce the occurrence rate of false positives, reduce influence, and improve The effect of accuracy
Pending Publication Date: 2020-05-15
PING AN TECH (SHENZHEN) CO LTD
View PDF0 Cites 2 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, in practical applications, many similar targets are often detected, resulting in many unnecessary false positives during image detection, which reduces the accuracy of detection.
In addition, there are often some deviations in the generation of ROI, which will also affect the detection results
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0051] It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.
[0052] The invention provides a target detection method. refer to figure 1 As shown, it is a schematic flowchart of a target detection method provided by an embodiment of the present invention. The method may be executed by a device, and the device may be implemented by software and / or hardware. In this embodiment, the device is an intelligent terminal.
[0053] In this embodiment, the target detection method includes:
[0054] S101. Build a detection target network structure, the detection target network structure includes a context-based ROI pooling layer, and the ROI pooling layer includes a first ROI pooling layer branch and a second ROI pooling layer branch;
[0055] S102, receiving an image of a target to be detected;
[0056] S103. Obtain a first ROI candidate frame of the image of the target to be detected t...
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
Login to View More Abstract
The invention relates to an image detection technology, and discloses a target detection method and a device, and a computer readable storage medium, the method comprises the following steps: establishing a detection target network structure comprising an ROI pooling layer based on a context relationship, the ROI pooling layer comprising a first ROI pooling layer branch and a second ROI pooling layer branch; receiving an image of a to-be-detected target; obtaining a first ROI candidate box of the image of the to-be-detected target through the first ROI pooling layer branch; acquiring a secondROI candidate box of the image of the to-be-detected target through a second ROI pooling layer branch, the size of the second ROI candidate box is larger than that of the first ROI candidate box, andthe second ROI candidate box comprises content in the first ROI candidate box and context information of the content in the first ROI candidate box; sending the content of the first ROI candidate boxand the content of the second ROI candidate box to a full connection layer of the detection target network structure; and updating the detection target network structure based on the full connection layer of the detection target network structure to complete image detection of the to-be-detected target.
Description
technical field [0001] The present invention relates to the field of image detection, in particular to an ROI multi-branch based target detection method, device and computer-readable storage medium. Background technique [0002] Images have the advantages of rich content, no language control, and easy international communication. In recent years, whether it is national defense, industrial manufacturing, news media, medical and health care or mass entertainment, all walks of life in society use image information more and more widely, so image detection is particularly important. [0003] In existing image detection tasks, most of the algorithms involved are calculated based on extracting a region of interest (ROI), which includes the target to be detected in the image detection task. However, in practical applications, many similar objects are often detected, resulting in many unnecessary false positives during image detection, which reduces the accuracy of detection. In ad...
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
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/32
CPCG06V10/25G06V2201/07
Inventor 章古月
Owner PING AN TECH (SHENZHEN) CO LTD