基于图像理解的铲车断齿无损检测算法设计毕业论文
2021-04-24 20:08:29
摘 要
在国内的矿山作业区常会用到一些大型挖掘器械,如挖掘机,装载机等。其中铲车在我国工业矿业生产中有着非常广泛的应用。但在铲车使用期间发生的铲齿断裂事故,不仅会降低矿山生产工作效率,还极易造成“过铁”事故,造成重大设备事故,严重影响矿山的安全生产,威胁操作人员的人身安全。故研发合适的断齿检测算法能够在铲齿发生脱落时,提供及时准确的警报信息,从而提高整个矿业生产的安全性和经济性。
本文以图像理解技术为依托在MATLAB中完成了相关检测算法的开发。具体内容如下。首先在对比传统的全局阈值分割算法OTSU和基于HSV色彩模型分割方法后,选择了根据H、S分量的阈值来对铲车图像进行图像二值化的方法。接下来通过结合面积阈值、离心率等形状特征,对二值化后的图像进行了去除杂质处理从而提取出了目标区域。最后根据同一位置两次采集的图片中目标区域的面积差值来判断是否铲齿发生断裂,从而完成了检测铲车断齿的目的。
通过在MATLAB中的多组实验,验证了算法的有效性,完成了检测铲车断齿的要求。同时在对实验结果进行分析后提出了相关的改进方向及方法
关键词:铲车铲齿;HSV色彩模型;图像理解;
Abstract
In the domestic mine operation area, some large excavation equipments such as excavators and loaders are often used. The forklift has a very wide range of applications in China's industrial mining production. However, the occurrence of shovel-tooth fracture during the use of forklifts will not only reduce the efficiency of mine production, but also easily lead to "over-iron" accidents, resulting in major equipment accidents, which will seriously affect the safe production of mines and threaten the personal safety of operators. Therefore, the development of a suitable broken tooth detection algorithm can provide timely and accurate alarm information when the scraping tooth falls off, thereby improving the safety and economy of the entire mining industry.
This article based on the image understanding technology in MATLAB completed the relevant detection algorithm development. The details are as follows. Firstly, after comparing the traditional global threshold segmentation algorithm OTSU and the HSV color model segmentation method, the method of binarizing image of forklift image based on threshold values of H and S components is selected. Next, by combining the shape features such as area threshold and eccentricity, the binarized image was subjected to impurity removal to extract the target area. Finally, according to the difference of the area of the target area in the two images collected at the same location, it is judged whether the teeth of the shovel have been broken, thereby completing the purpose of detecting the broken teeth of the shovel.
Through a series of experiments in MATLAB, the effectiveness of the algorithm is verified and the requirement of detecting the broken teeth of the forklift is completed. At the same time, after analyzing the experimental results, relevant improvement directions and methods were proposed.
Keywords: Forklift Shovel Tooth; HSV Color Model; Image Understanding
目录
摘要 I
Abstract II
第1章 绪论 1
1.1课题研究背景及意义 1
1.1.1 矿用挖掘机概述 1
1.1.2 课题研究背景 3
1.1.3课题研究意义 4
1.2国内外研究现状 4
1.3本文主要研究内容 5
第2章 基于图像理解技术的检测系统方案设计 6
2.1 典型图像处理系统概述 6
2.2 照明光源 6
2.3 工业相机 7
2.4光学镜头 8
2.5图像处理软件系统 9
2.6本章小结 10
第3章 应用图像理解技术检测铲车断齿的理论基础 11
3.1 图像处理概述及应用 11
3.2 彩色模型 11
3.2.1 RGB色彩模型 11
3.2.2 HSV色彩模型 12
3.3图像分割技术基础 13
3.4 本章小结 14
第4章 铲车断齿检测核心算法研究 15
4.1铲车铲齿脱落系统流程设计 15
4.2图像预处理 15
4.3图像二值化 16
4.3.1 基于最大类间方差法二值化图像 16
4.3.2 基于HSV色彩模型的二值化 16
4.4 铲车铲齿目标区域提取 21
4.4.1 零碎杂质的去除 21
4.4.2 标记相关区域 22
4.4.3 去除异常图像 22
4.4.4 基于离心率的杂质去除 23
4.5 铲车断齿检测 25
4.6章节小结 26
第5章 算法仿真及结果分析 27
5.1 仿真与分析 27
5.2 本章小结 29
第6章 总结与改进 30
6.1 全文总结 30
6.2 改进方向 30
参考文献 31
致谢 33
第1章 绪论
1.1课题研究背景及意义
1.1.1 矿用挖掘机概述
矿用挖掘机,又称电铲或绳铲,如图1.1所示是一种普遍用在公路、建筑、港口以及野外矿产中的挖掘装载设备,铲车的主要用途是铲装散状物料或者对土石方、矿石等进行简单的采掘操作。由于铲车具有作业效率高、设备灵活性好、相关控制技术成熟等优势,因此它在城市建设以及矿石采装工业中有着举足轻重的地位。
图1.1 矿用挖掘机(Pamp;H 4100)