Physical idea of the Retinex theory in color image enhancement and the influence of image quality in different intercepted region of image intensity
1
Citation
0
Reference
20
Related Paper
Citation Trend
Abstract:
The physical idea of Retinex theory that used in color image enhancement was revealed.The product of the convolution between a Gauss function and the original image,which was the smooth part of the original image,subtracted the original image in logarithm space.The left part was quick transform part of the original image.The details of the original image were highlighted.The sharper of the Gauss function was,the more details were highlighted.The smoother of the Gauss function was,the better hue of the image was showed.The result of multi-scale Retinex(MSR) has advantages in different single-scale Retinex(SSR).The method of standard deviation in intercepted region of handled image by MSR was researched.The result is obvious that the image quality intercepted in the region of [μ-σ,μ+σ],which μ is mean of image handled by MSR,is better than intercepted in the region of [μ-2σ,μ+2σ] and [μ-3σ,μ+3σ].Keywords:
Color Constancy
Hue
Convolution (computer science)
Cite
Retinex is a method used for image processing. Image processing has a great role in Medical science. Medical images such as MRI, CT, Ultrasound, X-Ray has to be processed for proper diagnosis. Retinex technique can be used for the processing of these images. By retinex processing it can provide better dynamic range compression, color consistency and lightness rendition. The different methods proposed by retinex algorithm includes Light Compensation Algorithm in Color Facial Image, Retinex for bridging the gap between color images and the human observation of scenes, Color Image Contrast Enhancement by Retinex, Color Image Enhancement with Adaptive Filter. In this paper we discuss about an algorithm for frequency domain based high resolution retinex for medical image processing. Initially we introduce Fast Fourier Transform to the image. Then Gaussian filtering is done and the inverse fourier transform is taken. Next step is to apply logarithmic function. Finally gain/offset is applied to obtain the enhanced output image. The method can produce good contrast enhancement. It can be used for both color image and gray image. Since it can process gray images the medical images can be processed successfully. The output for Single Scale Retinex, Multi Scale Retinex and Multi Scale Retinex with Color Restoration are obtained.
Color Constancy
Tone mapping
Homomorphic filtering
Cite
Citations (2)
Images are valuable information sources for many scientific and engineering applications. However, images captured in poor illumination conditions would have a large portion of dark regions that could heavily degrade the image quality. In order to improve the quality of such images, a restoration algorithm is developed here that transforms the low input brightness to a higher value using a modified Multi-Scale Retinex approach. The algorithm is further improved by a entropy based weighting with the input and the processed results to refine the necessary amplification at regions of low brightness. Moreover, fine details in the image are preserved by applying the Retinex principles to extract and then re-insert object edges to obtain an enhanced image. Results from experiments using low and normal illumination images have shown satisfactory performances with regard to the improvement in information contents and the mitigation of viewing artifacts.
Color Constancy
Cite
Citations (3)
Color Constancy
Color balance
Color histogram
RGB color model
RGB color space
Hue
Cite
Citations (49)
Retinex theory points that the color of each object is determined by its reflectance,and thus several Retinex-based algorithms are developed to uncover the real appearance of objects by estimating the reflectance of them in image. However,since they assume that components of the incidental lights on two different points are the same,the color distortion problem cannot be avoided. Now an improved Retinex algorithm was proposed to adjust each pixel by utilizing the difference between its original gray value and the one after Gaussian filtering. The simulation experiment and the two known evaluation criteria show that this algorithm's higher accuracy and better image enhancement performance than the traditional algorithm.
Color Constancy
Cite
Citations (3)
In order to improve the visibility of foggy images, this paper uses two models to iteratively refine the image. In the first model, the image is first enhanced by histogram equalization and then enhanced by the Retinex algorithm. In the second model, the image is firstly enhanced with the Retinex algorithm, and then the gamma correction is used to adjust the brightness. From a theoretical analysis and practical experiments, this method improves the sharpness of the image while enhancing the image detail information and restoring the image color.
Color Constancy
Visibility
Adaptive histogram equalization
Gamma correction
Cite
Citations (0)
For the multi-scale Retinex algorithm (MSR), the image contrast and image information is not ideal enough. In this paper, we propose an algorithm of image haze removal based on the fusion of dark prior and Retinex theory. Firstly, the algorithm of dark primary color is used to restore the haze image. Then, the multi-scale Retinex algorithm is applied to enhance the haze. Finally, the average value of brightness, standard deviation, entropy, mean square error, and peak signal-to-noise ratio were used as evaluation criteria for image enhancement. The simulation results of Matlab software show that the smog image processed by this algorithm has increased image contrast and more image information. We conduct experiments to prove that the proposed algorithm can provide a better representation.
Color Constancy
Haze
Peak signal-to-noise ratio
Cite
Citations (2)
With popular of Digital Still Camera-DSC, higher image quality is required. One of the subjects is that image quality at shadow area caused by the narrow dynamic range of the CCD devices is improved automatically. Conventionally, gamma transformation, histogram equalization, and etc. have been utilized for this improvement, but these are not always enough improvement. Recently, examinations applying to retinex theory taking into account of human eyes characteristics proposed by Land are paid attention. This algorithm renders image at shadow area clearly and effectively using spatial information between surrounding pixels arranged into two dimensions. Typical methods are Single-scale retinex(SSR) and Multi-Scale Retinex(MSR). These methods, however, does not always work on practical use in terms of color correction of the printed images with different RGB density distribution. In order to improve the issues of MSR, we propose the Modified Linear Multi-scale retinex (ML-MSR) method. A modified method consists of (a) linear computation processing and (b) synthesis both the original images and the images obtained by the linear MSR. By the simulation for the images printed by DSC, we show that ML-MSR can improve the visibility at shadow areas keeping with both the color balance and saturation, comparing with the conventional methods, such as histogram equalization and MSR proposed by Jobson. In general, a processing time of MSR remarkably increases with the size of Gaussian averaging filter to compute the weighted average. We describe about faster processing method of the ML-MSR algorithm, which has been shorten by using the thinning out of surrounding pixels and simplicity of average processing.
Color Constancy
RGB color model
Gamma correction
Tone mapping
Color balance
Cite
Citations (15)
In view of the problems of low image brightness, obvious noise, poor contrast, and difficulty in obtaining detailed information in dark areas under low light environment, we propose to combine the improved particle swarm optimization algorithm with a single-scale Retinex algorithm. We convert the original RGB image to the HSI color space,and each pixel of the low light image is classified separately. The adjacent pixels are calculated with the same kernel function value. The pixels with different H values are use different filter templates to complete the image enhancement. And solve the problem of image halo effect and color distortion caused by the real-time operation of the Retinex algorithm spatial filtering. Experimental results show that the proposed algorithm performs well in brightness, contrast, and color restoration.
Color Constancy
RGB color model
Color balance
Kernel (algebra)
Color histogram
Distortion (music)
Cite
Citations (7)
본 논문은 상대적으로 대비도 차이가 크게 나타나는 역광 이미지에 대해서 Retinex 알고리즘을 적용하여 보정 했을 경우 발생하는 밝은 영역에서의 컬러성분의 손실을 개선하기 위한 새로운 기법을 제안한다. 역광 이미지의 경우 밝은 영역과 어두운 영역에 대한 밝기 차이가 매우 크게 발생하기 때문에 Retinex 알고리즘을 이용하여 영상의 대비도를 향상시킬 경우 밝은 영역에서의 컬러 성분이 손실되는 현상이 발생한다. 이러한 손실을 보완하기 위해서 원본 영상의 밝은 영역에 해당하는 컬러 성분을 Retinex 알고리즘으로 보정된 영상에 추가해준다. K-mean 알고리즘을 이용하여 원본 영상에서의 밝은 영역, 어두운 영역, 중간 영역을 분리하고 밝은 영역에 대해서의 컬러 성분을 추가적으로 복원해 주며, 중간 영역에 대해서는 히스토그램에서의 위치를 기준으로 밝고 어두운 성분에 대한 비율을 고려하여 각 비율에 따라 원본 영상과 Retinex 복원 영상의 밝기 값을 함께 이용하도록 한다. 제안하는 알고리즘의 성능 평가를 위해 역광 현상이 강하게 나타나는 자연영상들을 대상으로 적용하여 기존의 Retinex 알고리즘보다 우수한 성능을 가지고 있음을 보였다. This paper proposes a new algorithm that improve color component of compensated image using Retinex method for back-light image. A back-light image has two regions, one of the region is too bright and the other one is too dark. If an back-light image is improved contrast using Retinex method, it loses color information in the part of brightness of the image. In order to make up loss information, proposed algorithm adds color components from original image. The histogram can be divided three parts that brightness, darkness, midway using K-mean (k=3) algorithm. For the brightness, it is used color information of the original image. For the darkness, it is converted using by Retinex method. The midway region is mixed between original image and Retinex result image in the ratio of histogram. The ratio is determined by distance from dark area. The proposed algorithm was tested on nature back-light images to evaluate performance, and the experimental result shows that proposed algorithm is more robust than original Retinex algorithm.
Color Constancy
Cite
Citations (2)
Retinex is a method used for image processing. Image processing has a great role in Medical science. Medical images such as MRI, CT, Ultrasound, X-Ray has to be processed for proper diagnosis. Retinex technique can be used for the processing of these images. By retinex processing it can provide better dynamic range compression, color consistency and lightness rendition. The different methods proposed by Retinex algorithm include Light Compensation Algorithm in Color Facial Image, Retinex for bridging the gap between color images and the human observation of scenes, Color Image Contrast Enhancement by Retinex, Color Image Enhancement with Adaptive Filter. Single scale Retinex causes halation due to Gaussian filter and it does not preserve the edges. While the multiscale Retinex has a very high computational cost. So in this project, we propose retinex algorithm based on wavelet transformation which has low computational cost, i.e. it takes a lesser amount of time and higher efficiency. The input image is processed by wavelet transform. Here the Gaussian filter and retinex is applied only to the half the resolution of the image. Thus the computational cost is reduced as the number of pixels for processing is reduced and the gaussian surround space can be small. Histogram equalization is applied to improve the visual effect. Moreover, we gain higher entropy by using clipping and gain/offset operation. At last, we compare the proposed output with that of the standard MSR output. The experimental results show that proposed method provides satisfactory image enhancement without halation.
Color Constancy
Cite
Citations (1)