A new perspective on decolorization: feature-preserving decolorization
- PDF / 2,234,337 Bytes
- 9 Pages / 595.276 x 790.866 pts Page_size
- 5 Downloads / 211 Views
ORIGINAL PAPER
A new perspective on decolorization: feature-preserving decolorization Orhan Akbulut1 Received: 20 June 2020 / Revised: 3 September 2020 / Accepted: 9 October 2020 © Springer-Verlag London Ltd., part of Springer Nature 2020
Abstract Information loss is a major problem in decolorization for color images. For example, the color-to-gray conversion may result in degradation in the contrast that affects visual perception quality. Preserving the source of information as much as possible is the main goal of the decolorization. This paper introduces an efficient contrast preservation method for decolorization based on spatial statistical distributions of the pixel-pair values. The statistical distributions are represented by the co-occurrence matrix in a compact form. Then, a feature extraction step is carried out by making use of this matrix. The feature extraction process is carried out for each channel of the color image and grayscale image to obtain source and target features. Featurepreserving criterion is constructed by l2 norm-based quality metric between source feature and target feature. The proposed method is remarkable because it is adaptable to any feature preserving, such as contrast. Moreover, there are no optimization phase, color space conversion, high complexity, and local mapping in the proposed method. Experimental results show that the performance of the proposed method is comparable to the existing decolorization approaches. Keywords Co-occurrence matrix · Contrast preserving · Decolorization · Feature preserving
1 Introduction Single-channel grayscale images are often preferred in image processing and computer vision fields because of their relatively low computational complexity compared to color images. Besides, it is more likely to use grayscale representations in digital printing and e-ink based applications. However, transforming color images into grayscale ones causes information loss and thus reduces the diversity of the image content. The information loss mainly stems from ignoring hue and saturation. The color-to-grayscale conversion process, so-called decolorization, can become crucial in grayscale-based processing such as image recognition [1,2]. To obtain feature descriptors, grayscale images are generally preferred due to their low computational complexity. This preference, however, may result in poor performance in the field of pattern-based image recognition.
B 1
Orhan Akbulut [email protected] Computer Engineering Department, Kocaeli University, Umuttepe Campus, 41001 Izmit/Kocaeli, Turkey
The baseline approach in decolorization is to assign fixed weights for each color channel, where the sum of the weights is equal to one. However, if the decolorization process is applied to all images with these weights, it may result in poor performance. An ideal decolorization process should meet certain requirements. The first one is to the preservation of the color contrast. Preserving color contrast in grayscale images can be critical for feature descriptors. The
Data Loading...
