Efficient Color Correction Using Normalized Singular Value for Duststorm Image Enhancement

A duststorm image has a reddish or yellowish color cast. Though a duststorm image and a hazy image are obtained using the same process, a hazy image has no color distortion as it has not been disturbed by particles, but a duststorm image has color distortion owing to an imbalance in the color channel, which is disturbed by sand particles. As a result, a duststorm image has a degraded color channel, which is rare in certain channels. Therefore, a color balance step is needed to enhance a duststorm image naturally. This study goes through two steps to improve a duststorm image. The first is a color balance step using singular value decomposition (SVD). The singular value shows the image’s diversity features such as contrast. A duststorm image has a distorted color channel and it has a different singular value on each color channel. In a low-contrast image, the singular value is low and vice versa. Therefore, if using the channel’s singular value, the color channels can be balanced. Because the color balanced image has a similar feature to the haze image, a dehazing step is needed to improve the balanced image. In general, the dark channel prior (DCP) is frequently applied in the dehazing step. However, the existing DCP method has a halo effect similar to an over-enhanced image due to a dark channel and a patch image. According to this point, this study proposes to adjustable DCP (ADCP). In the experiment results, the proposed method was superior to state-of-the-art methods both subjectively and objectively.

Efficient Sandstorm Image Enhancement Using the Normalized Eigenvalue and Adaptive Dark Channel Prior

A sandstorm image has features similar to those of a hazy image with regard to the obtaining process. However, the difference between a sand dust image and a hazy image is the color channel balance. In general, a hazy image has no color cast and has a balanced color channel with fog and dust. However, a sand dust image has a yellowish or reddish color cast due to sand particles, which cause the color channels to degrade. When the sand dust image is enhanced without color channel compensation, the improved image also has a new color cast. Therefore, to enhance the sandstorm image naturally without a color cast, the color channel compensation step is needed. Thus, to balance the degraded color channel, this paper proposes the color balance method using each color channel’s eigenvalue. The eigenvalue reflects the image’s features. The degraded image and the undegraded image have different eigenvalues on each color channel. Therefore, if using the eigenvalue of each color channel, the degraded image can be improved naturally and balanced. Due to the color-balanced image having the same features as the hazy image, this work, to improve the hazy image, uses dehazing methods such as the dark channel prior (DCP) method. However, because the ordinary DCP method has weak points, this work proposes a compensated dark channel prior and names it the adaptive DCP (ADCP) method. The proposed method is objectively and subjectively superior to existing methods when applied to various images.

Leveraging Deep Statistics for Underwater Image Enhancement

Underwater imaging often suffers from color cast and contrast degradation due to range-dependent medium absorption and light scattering. Introducing image statistics as prior has been proved to be an effective solution for underwater image enhancement. However, relative to the modal divergence of light propagation and underwater scenery, the existing methods are limited in representing the inherent statistics of underwater images resulting in color artifacts and haze residuals. To address this problem, this article proposes a convolutional neural network (CNN)-based framework to learn hierarchical statistical features related to color cast and contrast degradation and to leverage them for underwater image enhancement. Specifically, a pixel disruption strategy is first proposed to suppress intrinsic colors’ influence and facilitate modeling a unified statistical representation of underwater image. Then, considering the local variation of depth of field, two parallel sub-networks: Color Correction Network (CC-Net) and Contrast Enhancement Network (CE-Net) are presented. The CC-Net and CE-Net can generate pixel-wise color cast and transmission map and achieve spatial-varied color correction and contrast enhancement. Moreover, to address the issue of insufficient training data, an imaging model-based synthesis method that incorporates pixel disruption strategy is presented to generate underwater patches with global degradation consistency. Quantitative and subjective evaluations demonstrate that our proposed method achieves state-of-the-art performance.

A Marine Organism Detection Framework Based on the Joint Optimization of Image Enhancement and Object Detection

Underwater vision-based detection plays an increasingly important role in underwater security, ocean exploration and other fields. Due to the absorption and scattering effects of water on light, as well as the movement of the carrier, underwater images generally have problems such as noise pollution, color cast and motion blur, which seriously affect the performance of underwater vision-based detection. To address these problems, this study proposes an end-to-end marine organism detection framework that can jointly optimize the image enhancement and object detection. The framework uses a two-stage detection network with dynamic intersection over union (IoU) threshold as the backbone and adds an underwater image enhancement module (UIEM) composed of denoising, color correction and deblurring sub-modules to greatly improve the framework’s ability to deal with severely degraded underwater images. Meanwhile, a self-built dataset is introduced to pre-train the UIEM, so that the training of the entire framework can be performed end-to-end. The experimental results show that compared with the existing end-to-end models applied to marine organism detection, the detection precision of the proposed framework can improve by at least 6%, and the detection speed has not been significantly reduced, so that it can complete the high-precision real-time detection of marine organisms.

A Novel Color Recognition Model for Improvement on Color Differences in Products via Grey Relational Grade

ED light, a green energy-saving light source, can cause color cast. For this reason, LED light is seldom favored by designers. The purpose of the paper is to provide shoppers who are observing product colors in an LED-lighted setting with an innovative color identification model. Based on designers’ product color comparison, the paper employs high-reliability mechanic visual perception in combination with grey relational grade. Grey relational grade is applied to eliminate electrical fault pertaining to mechanic visual perception, whereby appropriate LED parameters and color cast inclination can be obtained. The paper first mimics retail store display windows. The color temperature and illuminance of LED light sources are adjustable. Two degrees of illuminance, including high illuminance (1500 lux) and low illuminance (500 lux), and two light source color temperatures, including yellow light (2700 K) and white light (4000 K), were assigned for study. Four colors, including red, yellow, blue and green of the natural color system, were selected as product colors. The mechanic visual perception sensor was used to identify the object (product) color, which is then converted into an RGB color model to serve as research data of color cast measurement, and the grey relational grade was applied to obtain the most appropriate LED light parameters and the color cast of the four colors. The data analysis reveals that green shows the least color cast when it is lighted by a yellow LED light source with low illuminance, yellow and blue have the least color cast when it is lighted by a white LED light source with high illuminance and red displays the least color cast when it is lighted by a white LED light source with low-illuminance. The analysis also indicates each color’s cast inclination in blackness, chromaticness and hue. As a result, LED light that is more acceptable to designers is suggested for display windows, thus reducing problems with product color cast.

A Unified Framework for Anomaly Detection of Satellite Images Based on Well-Designed Features and an Artificial Neural Network

Nowadays, as the number of remote sensing satellites launched and applied in China has been mounting, relevant institutions’ workload of processing raw satellite images to be distributed to users is also growing. However, due to factors such as extreme atmospheric conditions, diversification of on-board device status, data loss during transmission and algorithm issues of ground systems, defect of image quality is inevitable, including abnormal color, color cast, data missing, obvious stitching line between Charge-Coupled Devices (CCDs), and inconstant radiation values between CCDs. Product application has also been impeded. This study presents a unified framework based on well-designed features an Artificial Neural Network (ANN) to automatically identify defective images. Samples were collected to form the dataset for training and validation, systematic experiments designed to verify the effectiveness of the features, and the optimal network architecture of ANN determined. Moreover, an effective method was proposed to explain the inference of ANN based on local gradient approximation. The recall of our final model reached 81.18% and F1 score 80.13%, verifying the effectiveness of our method.

A Review on the Underwater Image Restoration and Enhancement Methods

Light scattering, as well as light absorption in the water, often cause underwater images to be hazy, poorly contrasted, and dominated by either green or blue colour cast. In this paper, we review some of the state-of-the-art approaches in which specifically designed to enhance the quality of the acquired images. These approaches are able to eliminate the color cast and haziness on the images as well as to improve the image colourfulness and contrast. The characteristics of each of the developed approaches are highlighted, and their performances are evaluated both subjectively and objectively by the quality assessment methods.

An improved K-means algorithm for underwater image background segmentation

Conventional algorithms fail to obtain satisfactory background segmentation results for underwater images. In this study, an improved K-means algorithm was developed for underwater image background segmentation to address the issue of improper K value determination and minimize the impact of initial centroid position of grayscale image during the gray level quantization of the conventional K-means algorithm. A total of 100 underwater images taken by an underwater robot were sampled to test the aforementioned algorithm in respect of background segmentation validity and time cost. The K value and initial centroid position of grayscale image were optimized. The results were compared to the other three existing algorithms, including the conventional K-means algorithm, the improved Otsu algorithm, and the Canny operator edge extraction method. The experimental results showed that the improved K-means underwater background segmentation algorithm could effectively segment the background of underwater images with a low color cast, low contrast, and blurred edges. Although its cost in time was higher than that of the other three algorithms, it none the less proved more efficient than the time-consuming manual segmentation method. The algorithm proposed in this paper could potentially be used in underwater environments for underwater background segmentation.