Abstract:Under low illumination, the colour constancy of human vision can be used for correctly determining the colour of objects according to the fixed reflection coefficient of external light and objects. However, video image acquisition equipment does not implement the colour constancy characteristic of the human visual system. Under low illumination, only a small amount of light is reflected from the surface of the imaged object; as a result, the captured image is underexposed. After statistical analysis of low-light images, these inverted underexposed images appear foggy. Inversion is a uniform and reversible operation that is performed on the entire image. Hereby, a method is proposed for resolving low-light images using conventional physical models. First, a low-light image is inverted for obtaining a foggy image. Subsequently, a pyramid-type dense residual block network and a dark channel prior K-means classification method are applied to the foggy image, to calculate the transmission and atmospheric light. Finally, the parameters obtained from this solution are incorporated into the low-light imaging model to obtain a clear image. We subjectively and qualitatively analysed the experimental results, and used information entropy and average gradient for objective quantitative analysis. We demonstrate that the algorithm improves the overall brightness and contrast of the imaged scenes, and the obtained enhanced images are clear and natural.

Spectral calculation model for machine vision image enhancement

Optimal Illumination for Visual Enhancement Based on Color Entropy Evaluation.

Enhancing the machine vision performance with multi-spectral light sources

Real-Color Image Denoised and Enhanced Synchronously Based on Wavelet Transform

Inferring Human Observer Spectral Sensitivities from Video Game Data.

Analyzing Color Light Additive Process Using Complex Frequency Spectrum

LED Light Design Method for High Contrast and Uniform Illumination Imaging in Machine Vision.

Low-light image enhancement algorithm based on an atmospheric physical model

A Virtual Color Space Based on Physical Model

A Learning Model Based On Pixels For Illumination Analysis

Joint RGB-Spectral Decomposition Model Guided Image Enhancement in Mobile Photography

Zero-LED: Zero-Reference Lighting Estimation Diffusion Model for Low-Light Image Enhancement

Spectral Optimization of White LED Based on Mesopic Luminance and Color Gamut Volume for Dim Lighting Conditions

Prediction model for indoor light environment brightness based on image metrics

Light optimal design for machine vision based on reflection

A New Color Constancy Model for Machine Vision

Optimal Illumination for Local Contrast Enhancement Based on the Human Visual System.

Optimising the Illumination Spectrum for Enhancing Tissue Visualisation

Image Dynamic-Range Enhancement Based on Human Visual Adaption Mechanism

Optimization Modulation Method for 3D Spectral Data Cube Using Linear Encoding of Intrinsic Chromatic Aberration

A Physiologically-Based Adaptive Three-Gaussian Function Model for Image Enhancement