{"title":"Citra Digital Untuk Klasifikasi Kualitas Udang Windu Menggunakan Algoritma GLCM dan K-Nearest Neighbor","authors":"Najirah Umar, Fiqri Haikal, M. Razak","doi":"10.31294/inf.v9i2.13686","DOIUrl":null,"url":null,"abstract":"Shrimp is a food that is easily damaged based on direct observation of the shrimp sorting process carried out by distributors or fishermen to select shrimp based on quality still using the manual method and sometimes the sorting results are still not in accordance with the quality of the shrimp and the quality indicators are only seen from the physical such as the weight or size of the shrimp, so that good quality shrimp can be mixed with less good quality, therefore contamination will occur which causes good quality shrimp to rot quickly. This final project aims to build an image processing system that applies the Gray-level Co-occurrence Matrix (GLCM) and K-nearest Neighbor (K-NN) algorithms to detect the quality level of Windu shrimp. The first process in this research is to perform image acquisition. That is, collecting several digital images of each quality of shrimp to use as an object. In addition, a pre-processing process is also carried out, namely changing the image to grayscale. Then the feature extraction process uses the Gray Level Co-occurrence Matrix (GLCM) method to obtain feature data from all digital images and classify them using the K-Nearest Neighbor (K-NN) method. The test results give an accuracy of 10 samples, it was found that as much as 80% got the results of quality classification information in accordance with the system. And this system is able to provide decision solutions in determining the quality classification of Windu Shrimp, while based on the results of blackbox testing, this system produces a percentage of application ease of use as much as 92%.","PeriodicalId":32029,"journal":{"name":"Proxies Jurnal Informatika","volume":"8 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2022-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proxies Jurnal Informatika","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.31294/inf.v9i2.13686","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Shrimp is a food that is easily damaged based on direct observation of the shrimp sorting process carried out by distributors or fishermen to select shrimp based on quality still using the manual method and sometimes the sorting results are still not in accordance with the quality of the shrimp and the quality indicators are only seen from the physical such as the weight or size of the shrimp, so that good quality shrimp can be mixed with less good quality, therefore contamination will occur which causes good quality shrimp to rot quickly. This final project aims to build an image processing system that applies the Gray-level Co-occurrence Matrix (GLCM) and K-nearest Neighbor (K-NN) algorithms to detect the quality level of Windu shrimp. The first process in this research is to perform image acquisition. That is, collecting several digital images of each quality of shrimp to use as an object. In addition, a pre-processing process is also carried out, namely changing the image to grayscale. Then the feature extraction process uses the Gray Level Co-occurrence Matrix (GLCM) method to obtain feature data from all digital images and classify them using the K-Nearest Neighbor (K-NN) method. The test results give an accuracy of 10 samples, it was found that as much as 80% got the results of quality classification information in accordance with the system. And this system is able to provide decision solutions in determining the quality classification of Windu Shrimp, while based on the results of blackbox testing, this system produces a percentage of application ease of use as much as 92%.