Open AccessIdentifikasi Kesegaran Ikan Nila menggunakan Teknik Citra Digital
Author(s) -
Juli Elprida Hutagalung,
Mhd Ihsan Pohan,
Yuli Happy Marpaung
Publication year - 2020
Publication title -
jurnal komputer dan informatika
Language(s) - English
Resource type - Journals
ISSN - 2722-4368
DOI - 10.53842/juki.v2i1.23
Subject(s) - grayscale , rgb color model , artificial intelligence , tilapia , computer vision , pixel , computer science , digital image , standard test image , image processing , image (mathematics) , histogram , process (computing) , mathematics , fish <actinopterygii> , fishery , biology , operating system
Fish contain many nutrients that are very beneficial for the body, but often fish are traded in a state of death as well as being alive. To observe the freshness of tilapia is done by the introduction of color changes that appear on digital images using the least squares method. The purpose of this research is to build an image management application system to detect the freshness of tilapia. The data used are 10 samples of tilapia images which are photographed every 1 hour for 15 hours and obtained 150 image data and then processed and analyzed using the least squares method. The first process begins with image processing by cropping at the edge of the eye of the original image and then proceed with resizing to 1000 x 1000 pixels and changing the image format to *. Png. After the image has been processed then the average value is calculated rata grayscale uses the 'rata_rata Gambar' application system and an equation is stored which is stored as training data on the application system. After the image has been processed then the image is input into the system, the image will be converted into grayscale form and displayed at a predetermined place together with the rgb and grayscale histograms and then calculated using the least squares method. The last process we do is matching the test image with the image stored as training data and we conclude whether the image is (very fresh, fresh, fresh enough, not fresh, or very not fresh), the percentage of freshness of the anchor fish, and the length of time the anchor fish dies. This study used 150 samples of fish images from fresh fish that were still very fresh until the fish were not very fresh (rotten).