# 基于机器视觉的Mini LED芯片传统缺陷检测

**Repository Path**: firecar20183210078/Defect-detection

## Basic Information

- **Project Name**: 基于机器视觉的Mini LED芯片传统缺陷检测
- **Description**: 基于机器视觉的传统缺陷检测, 即采用标准图片和待测图片进行pixel to pixel的XOR操作，这样可以得到瑕疵的位置, 得到瑕疵位置的像素点数, 这里判断是否缺陷只是简单的评估瑕疵的像素点数是否超过一定设置的阈值。 采用qt5实现了可视化界面操作
- **Primary Language**: C++
- **License**: GPL-3.0
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 8
- **Forks**: 6
- **Created**: 2021-12-31
- **Last Updated**: 2026-01-08

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# Traditional defect detection of mini LED chip based on machine vision



#### Introduction

The traditional defect detection based on machine vision, that is, the XOR operation of pixel to pixel is carried out by using the standard picture and the picture to be tested. In this way, the location of the defect and the number of pixels at the location of the defect can be obtained. Here, to judge whether the defect is just a simple evaluation of whether the number of pixels of the defect exceeds a certain threshold. The visual interface operation is realized by Qt5



#### Software architecture

C + + language + mingw64 compiler + opencv + Qt5





#### Function

1. Realize defect detection of mini LED chip

2. Label the mini led in a picture

3. The realization of visual interface is convenient to adjust the threshold and observe the detection results

4. It can detect the key areas of the chip



#### Instructions for use



1. The source code is placed in the SRC folder, the header file is in the headfile folder, and the UI file is placed in the UI

2. Directly run main Exe to run

3. Img folder contains the pictures to be tested and the template pictures used when the program is running



#### Remarks

The program may flash back inexplicably. If you need to modify it for other scenes, you need to adjust the size of the template picture