Multiscale Retinex

Abstract

While the retinex theory aimed at explaining human color perception, its derivations have led to efficient algorithms enhancing local image contrast, thus permitting among other features, to "see in the shadows". Among these derived algorithms, Multiscale Retinex is probably the most successful center-surround image filter. In this paper, we offer an analysis and implementation of Multiscale Retinex. We point out and resolve some ambiguities of the method. In particular, we show that the important color correction final step of the method can be seriously improved. This analysis permits to formulate an automatic implementation of Multiscale Retinex which is as faithful as possible to the one described in the original paper. Overall, this implementation delivers excellent results and confirms the validity of Multiscale Retinex for image color restoration and contrast enhancement. Nevertheless, while the method parameters can be fixed, we show that a crucial choice must be left to the user, depending on the lightning condition of the image: the method must either be applied to each color independently if a color balance is required, or to the luminance only if the goal is to achieve local contrast enhancement. Thus, we propose two slightly different algorithms to deal with both cases. Source CodeThe source code, the code documentation, and the online demo are accessible at the IPOL web page of this article 1 . In this link an implementation is available for download. Compilation and usage instruction are included in the README.txt file of the compressed archive. This software includes the implementations of algorithms potentially linkable to patents.

Color Consistency and Local Contrast Enhancement for a Mobile Image-Based Change Detection System

“…Third, we apply the gain/offset function (Eqation (3)) before the color processing function, in contrast to [7] where the order of the two operations is reversed (i.e., we use I Ret i (x) in Equation (6), instead of R i (x, c)). We found that this prevents from color conversion issues for images that contain saturated colors, compared to the original color processing scheme [7,20]. In Figure 4, we show an example of an image that contains pixels with saturated colors, for example, pixels of the blue balls with I R = 0 (Figure 4a).…”

Análise experimental de algoritmos de constância de cor e segmentação para detecção de mudas de plantas

“…Um exemplo de como nosso sistema de visão é relativo pode ser avaliado na ilusão de ótica da Figura 10, onde o quadrado A e D possuem a mesma intensidade. Além do estudo fisiológico, a teoria foi proposta para obtenção de uma análoga constância luminosa em sistemas de visão de máquina, caracterizando o Retinex como uma estratégia para processamento de imagem, gerando realce de luminosidade e cor para imagens digitais (AGARWAL et al, 2006;KOSCHAN;ABIDI, 2008;PETRO;SBERT;MOREL, 2014).…”

“…Fazendo uso deste operador gaussiano como função de cálculo de região normalizada, um primeiro método, conhecido como Single-Scale Retinex (SSR) pode ser explorado (RAHMAN; WOODELL, 2004;PETRO;SBERT;MOREL, 2014). Sua forma matemática, a ser estendida para todos os canais da imagem, é dada por:…”

“…• Ganho e deslocamento canônico: Em termos práticos, consiste na readequação da distribuição dos píxeis ao longo da imagem, sendo que píxeis posicionados nas extremidades são limitados a valores máximos e mínimos, e demais valores são reajustados de forma a ocupar ao máximo este intervalo (RAHMAN; WOODELL, 2004;PETRO;SBERT;MOREL, 2014). Sua formula é expressa como segue, sendo G e b os valores de ganho e deslocamento respectivamente:…”

Midway Video Equalization

“…They usually enhance the contrast in each pixel depending on the information of its local neighborhood, with the aim of isolating the effects on dark and bright regions. Local image contrast methods are usually based on human perception, like, e.g., the multiscale retinex method [9,13].…”