摘要
在大米品质检测中,大米图像的分割是其中的必要环节。常用otsu阈值分割算法实现大米图像分割的过程。但是传统一维otsu算法只是基于灰度值信息的分割效果不理想,而传统二维otsu算法的存在计算冗余,过程复杂、分割需要的时间长,为此提出了一种大米图像分割的pso优化二维otsu算法,通过粒子群优化算法实时更新粒子速度和位置,利用其寻优特性能够快速求解得到分割阈值,实现大米图像分割,减少otsu算法自适应阈值的冗余计算,提高运行速度。结果表明该分割算法在大米图像分割实验中,有效降低了二维otsu运算过程的复杂度,减少了图像阈值分割的时间,并且没有对取值结果造成影响,寻找的阈值准确,提高了阈值分割的效率,具有一定的可行性,为大米图像处理研究和粮食品质检测提供了参考。
Rice quality detection is necessary in rice image segmentation.Commonly used Otsu threshold segmentation algorithm to achieve rice image segmentation technology.However,the traditional one-dimensional Otsu algorithm is only based on the gray value information,which is not ideal.And the traditional two-dimensional Otsu algorithm has high computational complexity and needs a long time for segmentation.In this paper,a PSO optimized two-dimensional Otsu algorithm for rice image segmentation is proposed.it is used to update the particle speed and position in real-time,and the segmentation threshold can be obtained quickly by using its optimization characteristics,so as to realize rice image segmentation and reduce redundant calculation of adaptive threshold of Otsu algorithm high operating speed.The results show that the algorithm effectively debases the computational complexity of two-dimensional Otsu algorithm,reduces the time of image threshold segmentation,and the threshold value is accurate,which improves the efficiency of threshold segmentation,and has certain feasibility.It provides a reference for rice image processing research and grain quality detection.
作者
吕婧
杨红卫
Lv Jing;Yang Hong-wei(Key Laboratory of Grain Information Processing and Control(Henan University of Technology),Ministry of Education,Henan Zhengzhou 450001;College of Information Science and Engineering,Henan University of Technology,Henan Zhengzhou 450001)
出处
《电子质量》
2020年第11期143-147,共5页
Electronics Quality
基金
国家自然科学基金项目(U1404617)。
关键词
大米图像分割
粮食图像分析
OTSU算法
阈值分割
二维直方图
PSO算法
目标背景
全局最优
rice image segmentation
grain image analysis
OTSU algorithm
threshold segmentation
twodimensional histogram
PSO algorithm
target background
global optimization