Colorization is a process of converting grayscale images into visually acceptable color images. The main goal is to convince the viewer of the authenticity of the result. Grayscale images that need to be colorized are, in most cases, images with natural scenes. Over the last 20 years a wide range of colorization methods has been developed – from algorithmically simple, yet time- and energy-consuming because of unavoidable human intervention to more complicated, but simultaneously more automated methods. Automatic conversion has become a challenging area that combines machine learning and deep learning with art. This paper presents an overview and evaluation of grayscale image colorization methods and techniques applied to natural images. The paper provides a classification of existing colorization methods, explains the principles on which they are based, and highlights their advantages and disadvantages. Special attention is paid to deep learning methods. Relevant methods are compared in terms of image quality and processing time. Different metrics for color image quality assessment are used. Measuring the perceived quality of a color image is challenging due to the complexity of the human visual system. Multiple metrics used to evaluate colorization methods provide results by determining the difference between the predicted color value and the ground truth, which in several cases is not in coherence with image plausibility. The results show that user-guided neural networks are the most promising category for colorization because they successfully combine human intervention and neural network automation.

Decolorization is an image processing technique that converts a color input image into a grayscale image. This paper discusses the decolorization process and provides an overview of the methods based on the different principles used: basic conversion from RGB to YUV format using ITU Recommendations 601, 709, and 2020; basic conversion from RGB to LAB color space; the method using cumulative distribution function of color channels; one global decolorization method; and one based on deep learning. The grayscale images produced by these methods were evaluated using four objective metrics, allowing for a thorough analysis and comparison of the decolorization results. Additionally, the execution speed of the algorithms was assessed, providing insight into their performance efficiency. The results demonstrate that different metrics evaluate the decolorization methods differently, highlighting the importance of selecting an appropriate metric that aligns with the subsequent image processing tasks following decolorization. Furthermore, it was shown that the decolorization methods depend on the content of the images, performing better on natural images than on artificially generated ones. The decolorization methods were also examined in the context of object segmentation and edge detection. The results from segmentation and edge detection were aligned with the decolorization results, revealing that certain objective metrics for evaluating decolorization more effectively assessed the properties of the decolorized images, which are crucial for successful object segmentation and edge detection.

Assessing the quality of colorized images remains challenging as most colorization artifacts arise from high-level semantic errors in the form of implausible or unnatural color assignments, rather than conventional low-level distortions such as noise, blur, or compression artifacts. The evaluation process is further complicated by the inherently subjective nature of color perception and the difficulty of accurately modeling human responses to color. Motivated by the limitations of current image quality assessment metrics for this task, we propose TRIPSI (Triple-Source Realigned Integrated Perceptual Semantic Index), a hybrid full-reference framework which approaches colorization quality assessment as a pattern recognition problem in a learned semantic-aware feature space. TRIPSI fuses three complementary deep pre-trained models, TOPIQ, LIQE, and DreamSim, into a unified framework by applying rank normalization to individual model scores per dataset to ensure comparability across varying output scales before aggregating the scores with equal weights. LIQE captures explicit color distortions. TOPIQ focuses on semantically important regions and color saturation artifacts. DreamSim measures color-preserving pattern agreement between deep feature representations of a colorized image and its ground-truth reference color image in a learned semantic-aware embedding space. By explicitly incorporating color-aware semantic representations at multiple levels, results across multiple datasets show that TRIPSI closely reflects human perceptual judgments, highlighting the effectiveness of semantic pattern modeling for quality assessment in image colorization.