Color segmentation in multicolor images using node‐growing self‐organizing map

Abstract

This article presents a clustering algorithm based on node‐growing self‐organizing map (NGSOM) to classify colors on color images automatically and accurately partition the regions of different colors for color measurement. Based on the CIEDE2000 criterion, pixels in a multicolor image are grouped into a number of visually distinguishable color regions in which pixel distribution information is provided as the input of the NGSOM network for further segmentation. As an unsupervised clustering algorithm, the NGSOM randomly selects two initial nodes from the input source without a predefined network structure and grows nodes according to color differences between the first best matching unit (FBMU) and the second best matching unit (SBMU) of the current input data. Unlike a traditional SOM, the NGSOM utilizes learning rates varying with the pixel distributions of major color regions. The node‐growing procedure is terminated when all the input data are examined. Compared with some commonly used color clustering algorithms, the proposed algorithm possesses a better peak signal‐to‐noise ratio (PSNR) and higher time efficiency. The NGSOM can be used for a wide range of applications, including fabric colorfastness assessment, painting conservation, and scenic identification in aerial images.

Pattern‐driven color pattern recognition for printed fabric motif design

“…The color image processing algorithms may be sensitive to the modeling way of color space. 18 Color image processing method based on low-level feature is usually performed on one dimensional color channel. It is necessary to preserve the desired color features when encoding the colors by reducing the number of color channels in RGB image.…”

“…There are many methods for color pattern segmentation. 18,[33][34][35][36] We propose a multi-region fuzzy competition segmentation method to segment the selected or matched image patches. The reason is that the proposed method is formulated in a variational framework, 37 which is an unsupervised segmentation process and can be solved by fast algorithms.…”

Multi-color measurement of printed fabric using the hyperspectral imaging system

“…Our algorithm and the other segmentation algorithms were also compared for time complexity (Figure 11). This involves determining the sum of time searching for the optimal number of clusters (2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15) and the time required to calculate the SI values. These results show the time taken on the algorithms while tested on the above five printed fabrics, showing that our algorithm performs more efficiently than other segmentation algorithms.…”

“…[9][10][11] With the development of clustering technology, many reasonable algorithms have been proposed for image segmentation of printed fabrics, which can automatically classify colors and accurately partition the regions of different colors for measurement without needing to identify the number of colors. [12][13][14][15][16][17] However, further improvements are necessary to ensure efficient execution of these algorithms.…”

Automatic COVID-19 lung infected region segmentation and measurement using CT-scans images

“…Various methodologies in computer vision have been proposed to deal with different sides to combat the COVID-19 pandemic, including segmentation and classification methods [4] . These approaches can be classified into two fundamental classes: Classical Machine Learning and Deep Learning methods [5] .…”