Analysis of rice granules using image processing and neural network
Analysis of rice granules using image processing and neural network
复制标题
使用图像处理和神经网络分析大米颗粒
DOI：
10.1109/cict.2013.6558219
复制
发表时间：
2013
期刊：
影响因子：
--
通讯作者：
S. Rubalya Valantina
中科院分区：
文献类型：
--
作者：
Periasamy Neelamegam;S. Abirami;K. Vishnu Priya;S. Rubalya Valantina
在食品加工业中，颗粒状食品材料的分级是必要的，因为材料样品会受到掺假。在过去，颗粒或颗粒形式的食品通过筛子或其他机械装置用于分级目的。本文对印度香米颗粒进行了分析，利用图像处理和神经网络对颗粒进行分类分级，以评价其性能。数字成像技术是一种非接触式提取稻米颗粒特征的有效方法。使用相机获取水稻的图像。灰度转换、中值平滑、自适应阈值、Canny边缘检测、Sobel边缘检测、形态学操作、定量信息提取是使用图像处理技术通过开源计算机视觉（Open CV）对所获取的图像执行的检查，所述开源计算机视觉（Open CV）是辅助真实的图像处理的函数库。从图像中获取的形态特征被赋予神经网络。这项工作已经完成，以确定相关的质量类别为基础的参数为一个给定的大米样品。图像处理的性能大大缩短了运算时间，提高了作物识别率。与人类专家的输出相比，神经网络系统获得的分级结果显示出更高的准确性。
In food handling industry, grading of granular food materials is necessary because samples of material are subjected to adulteration. In the past, food products in the form of particles or granules were passed through sieves or other mechanical means for grading purposes. In this paper, analysis is performed on basmati rice granules; to evaluate the performance using image processing and Neural Network is implemented based on the features extracted from rice granules for classification grades of granules. Digital imaging is recognized as an efficient technique, to extract the features from rice granules in a non-contact manner. Images are acquired for rice using camera. Conversion to gray scale, Median smoothing, Adaptive thresholding, Canny edge detection, Sobel edge Detection, morphological operations, extraction of quantitative information are the checks that are performed on the acquired image using image processing technique through Open source Computer Vision (Open CV) which is a library of functions that aids image processing in real time. The morphological features acquired from the image are given to Neural Network. This work has been done to identify the relevant quality category for a given rice sample based on its parameters. The performance of image processing reduced the time of operation and improved the crop recognition greatly. Grading results obtained from Neural Network system shows greater accuracy when compared with the outputs from human experts.