Grayscale Image Colorization Research Articles

Affective Image Colorization

Colorization of gray-scale images has attracted many attentions for a long time. An important role of image color is the conveyer of emotions (through color themes). The colorization with an undesired color theme is less useful, even it is semantically correct. However this has been rarely considered. Automatic colorization respecting both the semantics and the emotions is undoubtedly a challenge. In this paper, we propose a complete system for affective image colorization. We only need the user to assist object segmentation along with text labels and an affective word. First, the text labels along with other object characters are jointly used to filter the internet images to give each object a set of semantically correct reference images. Second, we select a set of color themes according to the affective word based on art theories. With these themes, a generic algorithm is used to select the best reference for each object, balancing various requirements. Finally, we propose a hybrid texture synthesis approach for colorization. To the best of our knowledge, it is the first system which is able to efficiently colorize a gray-scale image semantically by an emotionally controllable fashion. Our experiments show the effectiveness of our system, especially the benefit compared with the previous Markov random field (MRF) based method.

Automatic grayscale image colorization using histogram regression

Colorization aims to adding colors to a grayscale image. This task is ill-posed in the sense that assigning the colors to a grayscale image without any prior knowledge is ambiguous. Most of the previous methods require some amount of user interventions, making colorization a hard work. Motivated by this, a novel automatic grayscale image colorization method based on histogram regression is presented in this paper. A source image is adopted to provide the color information. Locally weighted regression is performed on both the grayscale image and the source image. Thus, the feature distributions of two images can be obtained. Then, a new matching method is proposed to align these features by finding and adjusting the zero-points of the histogram. When the luminance-color correspondence was achieved, the grayscale image is colorized in a weighted way. Moreover, a new evaluation method is specially designed to assess the confidence of the colorization results. Various experiment results are given to show the validity of this method.

Colorization of Gray Scale Images in YCbCr Color Space Using Texture Extraction and Luminance Mapping

Gray scale image colorization is a useful application in the world of image processing. This work presents a general technique for “colorizing” grayscale images by transferring color between a source or color image and a destination or target or grayscale image by matching luminance and texture information between the images. This technique of the gray image coloring uses a very simple algorithm that chooses a decorrelated color space and then applies simple operations there. Thus the color space with decorrelated axes is a useful tool for manipulating color images in this work. Colorization of a gray scale image is achieved here by matching texture features of gray image with texture features of windows of colored image and then by imposing mean and standard deviation onto the data points in a simple operation, thus believable output images are obtained given suitable input images. Here, we use texture features like energy, entropy, homogeneity, contrast and correlation based on correlation matrix for the purpose of texture matching between two images.

Automated Colorization of Grayscale Images Using Texture Descriptors and a Modified Fuzzy C-Means Clustering

A novel example-based process for Automated Colorization of grayscale images using Texture Descriptors (ACTD) without any human intervention is proposed. By analyzing a set of sample color images, coherent regions of homogeneous textures are extracted. A multi-channel filtering technique is used for texture-based image segmentation, combined with a modified Fuzzy C-means (FCM) clustering algorithm. This modified FCM clustering algorithm includes both the local spatial information from neighboring pixels, and the spatial Euclidian distance to the cluster’s center of gravity. For each area of interest, state-of-the-art texture descriptors are then computed and stored, along with corresponding color information. These texture descriptors and the color information are used for colorization of a grayscale image with similar textures. Given a grayscale image to be colorized, the segmentation and feature extraction processes are repeated. The texture descriptors are used to perform Content-Based Image Retrieval (CBIR). The colorization process is performed by Chroma replacement. This research finds numerous applications, ranging from classic film restoration and enhancement, to adding valuable information into medical and satellite imaging. Also, this can be used to enhance the detection of objects from x-ray images at the airports.

A new image encryption algorithm based on chaos

In this letter, we introduce a new image encryption algorithm based on iterating chaotic maps. Using the pseudorandom sequence generated by a group of one dimensional chaotic maps, the proposed algorithm realizes fast encryption and decryption of both gray-scale image and true color image. Moreover, the rounds of encryption could be set by the user. Theoretical analysis and numerical simulation prove the proposed algorithm effective and secure.

A New Colorization Algorithm Based on Dynamic Neighbor-Relevant Chromatic Channel Information

This paper is concerned with the problem of automatic colorization of grayscale images via color transfer. Some deficiencies of Welsh's colorization algorithm are pointed out and analyzed. Then, a new colorization algorithm is proposed, which uses a new similarity function and a new search strategy for pixel matching. The new function incorporates both the luminance information and chromatic values of previously colorized neighboring pixels. The intrinsic space clustering property of an image is taken into account in the strategy by using Hilbert filling curve scanning order. Experimental results showed that the proposed algorithm outperforms Welsh's algorithm in both colorization effect and computational complexity. Moreover, the proposed similarity function and search strategy can easily be embedded into other colorization algorithms to improve their performance.

Optimization based grayscale image colorization

An optimization based interactive grayscale image colorization method is presented in this paper. It is an interactive colorization method, whereas the only thing user need to do is to provide some color hints by scribbling or seed pixels. The main contribution of this paper is that the colorization method greatly reduces computation time with the same good results in image quality by quadtree decomposition based non-uniform sampling. Moreover, by introducing a new simple weighting function to represent intensity similarity in the cost function, annoying color diffusion among different regions is alleviated. Experiments show that this method gives the same good quality of colorized images as the method of Levin et al. with a fraction of the computational cost.

Patch - based Colorization of Grayscale Images

Patch - based Colorization of Grayscale Images