Multi-scale image target detection method, system, electronic equipment and storage medium
A technology of image target and detection method, which is applied in the field of image processing, can solve the problems of time-consuming and laborious collection of data sets, the inability to guarantee the multi-scale images of the camera, and the inability to guarantee the accuracy, etc., to increase the amount of data, expand the scale, and improve detection The effect of precision
Active Publication Date: 2021-07-30
WUHAN OPTICS VALLEY INFORMATION TECH +1
View PDF5 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, for the target detection task of a project, data is usually collected in the form of video streams. The collection of the data set is a time-consuming and laborious work, and for the target detection of the project, there is no guarantee that the multi-scale camera can be collected. image, that is, the accuracy of object detection in the image cannot be guaranteed when the focal length of the camera changes
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
[0029] The specific implementation manners of the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.
[0030] figure 1A multi-scale image target detection method provided by the present invention is applicable to the recognition of objects in images of any scale. The method includes: S1, input multiple small-scale images into a preset detection model, and obtain the preset detection The first object target frame in each small-scale image recognized by the model, the preset detection model is obtained by training according to the small-scale image training set; S2, the pixel coordinates of the first object target frame in each small-scale image Mapping to the large-scale image, obtaining the pixel coordinates of the second object target frame of the same object in the large...
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 provides a multi-scale image target detection method, system, electronic equipment and storage medium. A detection model is trained based on small-scale images, and the detection model is used to identify object target frames in small-scale images. The identified target frame of the object is mapped to the large-scale image, the target frame of the same object is found in the large-scale image, and finally the detection model is optimized by using different scale images, which not only increases the data of the detection model training set It also expands the scale of the images in the training set, and the optimized detection model is suitable for identifying objects in images of various scales, improving the accuracy of target detection.
Description
technical field [0001] The present invention relates to the field of image processing, and more specifically, to a multi-scale image target detection method, system, electronic equipment and storage medium. Background technique [0002] In recent years, with the continuous development and optimization of the target detection field, its detection accuracy has met the requirements, and with the concept of smart city, its rich information perception network provides a basic platform for target detection, and based on the perception network Different sensors form multi-view, multi-scale, and multi-resolution detection data. [0003] The current best way to achieve target detection is to use deep learning technology, such as Yolov5 target detection, and deep learning technology greatly depends on the size of the data set. Theoretically, when other conditions are the same, the larger the dataset, the higher the accuracy of the target detection model. However, for the target dete...
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 Patents(China)
IPC IPC(8): G06K9/00G06K9/32G06K9/46G06K9/62G06T7/73G06T7/80
CPCG06T7/80G06T7/73G06T2207/10016G06V20/10G06V10/25G06V10/462G06V2201/07G06F18/24
Inventor 洪勇晏世武吴培桐罗书培李江张翔
Owner WUHAN OPTICS VALLEY INFORMATION TECH