Color Cast Detection Research Articles

Primary Colour Detection in An Image

Primary Colour Detection in an Image, Traditional methods of color cast detection do not discriminate between images with true cast and those with dominant colors. This may result in an inaccuracy of the color cast measurement. In order to overcome the limitation of traditional methods, an approach based on image semantic is present. It can improve the accuracy and reliability of the detecting results by the means of recognizing and removing the dominant color objects · and analyzing the color distribution of the whole image. Class specific color detection is already implemented in some systems, but image classification usually relies on global image features only. The method of recognizing the dominant color objects is based on block-features and region-features by using K-Means Clustering in this project. It shows a significant improvement over previously achieved classification for a variety of critical image classes. The comparison of the results of the approach proposed to that of people shows that it is reliable and effective. This project proposes the use of Machine Learning Techniques and Deep Learning Algorithms.

Model of Image Color Difference and Partial Based On RGB Color Distribution Measuring

According to the color difference fabric dyeing measurement in production, partial color existed bad real-time problems. Color difference and color of RGB fabric images put forward by using the method of color distribution measurement of general, which needed rapid measurement of large area color. According to RGB distribution histogram of the image, the measured color with sample color similarity and distribution curve of characteristic value was calculated, obtained the RGB color model, and gives a method of a partial color judging. The experiment proves the feasibility of this measurement fabric color difference and the judgment method and effectiveness of partial color. Color cast detection method has overcome the limitations of the traditional methods of detecting image color cast. In order to partial color image detection is performed after correction, using a combination of gray world and a perfect reflection of the color correction method, to make up for the shortcomings of traditional method. The experimental results show that, the color deviation correction method of image analysis based on detection improved the reliability of color cast detection. The method uses characteristics of image analysis, it has universal applicability.

A generalized white-patch model for fast color cast detection in natural images

We present a generalized white-patch technique able to rapidly detect color cast of natural images. Instead of relying on the chromatic information of a single perfectly reflective patch in the image, as pure white-patch models do, we consider a connected region of pixels that will serve as white reference for the method. The pixels belonging to the white reference region must comply with three properties: 1) they do not have to be completely saturated; 2) they must belong to the p% of pixels with brightest intensity in the whole image (where p is a parameter of the model); 3) the area of the connected region formed by these pix-els must overcome a threshold of significance A (a second parameter). Color cast is detected if the average intensity in the three separated chromatic channels RGB is distant enough from a neutral grey level, where the distance is measured through an angular metric.

Color Correction with Zone System for Color Image

Color correction is a part of image processing to solve the problem of color cast which a particular and unwanted color dominates an image. The Zone system is a visual ruler that allows photographers to determine optimal film exposure and development. It is divided the full tonal levels into eleven zones ranging from zone 0 (pure black) to zone X (pure white), with zone V representing subjective middle gray which is corresponding to 18% reflectance. In this paper, we described a global color correction method with Zone system to remove the color cast from the color images. The proposed method included color cast detection and color correction two steps. In color cast detection step, we divided the histograms of chrominance components (Cb and Cr) in YCbCr color space into eleven zones. The probable adjusted pixels of color cast will be detected after the Cb and Cr values of each pixel compared with zone V. In color correction step, the adjusted image will be obtained when the RGB channel individually times gains that are derived from the mean values and minimal values of the adjusted pixels that are the top 90% of the probable adjusted pixels. Experimental results exhibited the proposed method can correct the color from the color cast. Finally, the proposed method was compared with other color correction methods in subjective and objective evaluations.

Research on color constancy under open illumination conditions

Color information plays a key role in the research fields of object recognition and image retrieval. However, the actual color varies by the conditions of illumination, especially the open natural daylight. Four different color constancy schemes are proposed in the paper to minimize the effects of open illumination conditions. (1) The color constancy scheme based on the image statistics is proposed, which includes the color cast detection and removal. (2) The color constancy scheme based on the color temperature curve is proposed, which combines Gaussian model with linear fitting to estimate color temperature curve. (3) The color constancy scheme based on the double exposure theory is proposed, which is able to reproduce a color image under typical illumination. (4) According to the concepts of supervised learning, the supervised color constancy scheme is proposed. The transformation of color values from unknown illumination to typical illumination is solved by improved Support Vector Regression (SVR).