Method for detecting and positioning target surface impact point based on convolutional neural network

A technology of convolutional neural network and positioning method, which is applied in the field of image detection and recognition, can solve the problems that the coordinate position of the shooting point cannot be accurately captured, and the positioning of the impact point on the target surface is not accurate enough, so as to reduce the amount of calculation and improve the accuracy of coordinates Effect
CN114119720APending Publication Date: 2022-03-01工学智能科技(合肥)有限公司
0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Current Assignee / Owner
工学智能科技(合肥)有限公司
Publication Date
2022-03-01

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention discloses a target surface impact point detection and positioning method based on a convolutional neural network, and belongs to the field of image detection. The objective of the invention is to more accurately position the central dot coordinate position of an impact point on a target surface and improve the resolution and precision of target scoring. The detection and positioning method comprises the following steps: firstly, acquiring a target surface image through a camera, processing the target surface image to obtain a target image expressed by a pixel point array, then performing brightness and color difference judgment on the target image, and constructing a rectangular frame for the pixel points passing judgment to obtain a candidate rectangular frame set; and processing the candidate rectangular frame set by adopting a convolutional neural network full connection layer to obtain a group of feature vectors, performing judgment through a classifier, and finally obtaining an accurate position of a target surface impact point by utilizing a regression device.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention relates to the field of image detection and recognition, in particular to a method for detecting and locating an impact point on a target surface based on a convolutional neural network. Background technique

[0002] With the rapid development of science and technology, the development of image processing technology has also entered the fast lane, and it is inseparable from people's daily life. It has played an increasingly important role in travel services, consumer services, security, medical care and other fields. The detection of a certain target on the image is an indispensable part of the image processing field. How to quickly and effectively detect the expected target and accurately locate it is a problem that technicians in this field need to solve.

[0003] The convolutional neural network target detection algorithm based on deep learning has the advantage of automatically learning and extracting the key features of the target, w...

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