Kmeans Clustering Segmentation on Water Microbial Image with Color and Texture Feature Extraction

Abstract

Image segmentation is one of the analytical processes for digital image recognition, where this process divides the digital image into several unique regions based on homogeneous pixels. The process of homogeneous grouping images is based on several colour, texture and shape features. Colour in digital image processing is very important because colour has many information humans can easily understand. Colour has various features, combining colour intensity and grey (grayscale) and binary (black and white) values. However, the colour feature extraction process has many weaknesses. If the object used has a very small si[1]ze and range area, the use of colour features needs to be combined with extraction, and the segmentation process can be maximized. This study uses colour and texture features in the extraction process. It uses bacterial objects (microbes) from water, with limited image quality and objects that tend to be difficult to identify. The colour space feature extraction process is combined with a Gabor filter so that the segmentation process produces high-quality accuracy. Good. The Gabor filter used in this study is combined with the L*a*b space vector to increase accuracy in the segmentation process. The results showed that the use of texture features resulted in an increase in accuracy of 17.5% by testing the cluster value of 1.2.

Analyzing the color availability of AI‐generated posters based on K‐means clustering: 74% orange, 38% cyan, 32% yellow, and 28% blue‐cyan

“…The number of clusters used in the algorithm determines the number of key colors extracted. [59][60][61] The core idea of the K-means clustering algorithm is to divide a given dataset N into k clusters, where the dataset N contains n data (n is a natural number), and find the clustering centres C 1 , C 2 , C 3 , …, C k of these clusters so that the sum of squared distances of each data point in the cluster and the clustering centre of the cluster is minimized as follows:…”

Selection of Optimal Segmentation Algorithm for Satellite Images by Intuitionistic Fuzzy PROMETHEE Method

“…It enables the discernment of many shades of color information, making color-based segmentation a more effective method for extracting information compared to intensity or texture-based segmentation [3]. The authors of [4] additionally mention that the process of extracting color features has its limitations, requiring a combination with other extraction methods to enhance segmentation accuracy, specifically if the object has a very small size and range area.…”

Enhancing CNNs Performance on Object Recognition Tasks with Gabor Initialization

“…Gabor filters have been successfully applied in computer vision for various tasks, such as recognizing objects, textures, and shapes. They have been used in tasks such as invariant object recognition [1], building and road structure detection from satellite images [1], license plate detection [2], traffic sign recognition [3], diagnosis of invasive ductal carcinoma of the breast [4], edge detection [5], texture segmentation [5], image classification [5], fingerprint and face recognition [5], texture recognition [6], and hyperspectral image classification [7]. Gabor filters are known for their ability to extract essential activations, their multi-orientation and multi-scale analysis capabilities, and their effectiveness in texture classification and feature extraction [3,4,7].…”