An Efficient Attentional Image Dehazing Deep Network Using Two Color Space (ADMC2-net).

Image dehazing has become a crucial prerequisite for most outdoor computer applications. The majority of existing dehazing models can achieve the haze removal problem. However, they fail to preserve colors and fine details. Addressing this problem, we introduce a novel high-performing attention-based dehazing model (ADMC2-net)that successfully incorporates both RGB and HSV color spaces to maintain color properties. This model consists of two parallel densely connected sub-models (RGB and HSV) followed by a new efficient attention module. This attention module comprises pixel-attention and channel-attention mechanisms to get more haze-relevant features. Experimental results analyses can validate that our proposed model (ADMC2-net) can achieve superior results on synthetic and real-world datasets and outperform most of state-of-the-art methods.

Development of a Cloud-Based Image Processing Health Checkup System for Multi-Item Urine Analysis.

With the busy pace of modern life, an increasing number of people are afflicted by lifestyle diseases. Going directly to the hospital for medical checks is not only time-consuming but also costly. Fortunately, the emergence of rapid tests has alleviated this burden. Accurately interpreting test results is extremely important; misinterpreting the results of rapid tests could lead to delayed medical treatment. Given that URS-10 serve as a rapid test capable of detecting 10 distinct parameters in urine samples, the results of assessing these parameters can offer insights into the subject's physiological condition. These parameters encompass aspects such as metabolism, renal function, diabetes, urinary tract disorders, hemolytic diseases, and acid-base balance, among others. Although the operational procedure is straightforward, the variegated color changes exhibited in the outcomes of individual parameters render it challenging for lay users to deduce causal factors solely from color variations. Moreover, potential misinterpretations could arise due to visual discrepancies. In this study, we successfully developed a cloud-based health checkup system that can be used in an indoor environment. The system is used by placing a URS-10 test strip on a colorimetric board developed for this study, then using a smartphone application to take images which are uploaded to a server for cloud computing. Finally, the interpretation results are stored in the cloud and sent back to the smartphone to be checked by the user. Furthermore, to confirm whether the color calibration technology can eliminate color differences between different cameras, and also whether the colorimetric board and the urine test strips can perform color comparisons correctly in different light intensity environments, indoor environments that could simulate a specific light intensity were established for testing purposes. When comparing the experimental results to real test strips, only two groups failed to reach an identification success rate of 100%, and in both of these cases the success rate reached 95%. The experimental results confirmed that the system developed in this study was able to eliminate color differences between camera devices and could be used without special technical requirements or training.

Low Light Image Enhancement Algorithm Based on Detail Prediction and Attention Mechanism.

Most LLIE algorithms focus solely on enhancing the brightness of the image and ignore the extraction of image details, leading to losing much of the information that reflects the semantics of the image, losing the edges, textures, and shape features, resulting in image distortion. In this paper, the DELLIE algorithm is proposed, an algorithmic framework with deep learning as the central premise that focuses on the extraction and fusion of image detail features. Unlike existing methods, basic enhancement preprocessing is performed first, and then the detail enhancement components are obtained by using the proposed detail component prediction model. Then, the V-channel is decomposed into a reflectance map and an illumination map by proposed decomposition network, where the enhancement component is used to enhance the reflectance map. Then, the S and H channels are nonlinearly constrained using an improved adaptive loss function, while the attention mechanism is introduced into the algorithm proposed in this paper. Finally, the three channels are fused to obtain the final enhancement effect. The experimental results show that, compared with the current mainstream LLIE algorithm, the DELLIE algorithm proposed in this paper can extract and recover the image detail information well while improving the luminance, and the PSNR, SSIM, and NIQE are optimized by 1.85%, 4.00%, and 2.43% on average on recognized datasets.

Retinex-Based Fast Algorithm for Low-Light Image Enhancement.

We proposed the Retinex-based fast algorithm (RBFA) to achieve low-light image enhancement in this paper, which can restore information that is covered by low illuminance. The proposed algorithm consists of the following parts. Firstly, we convert the low-light image from the RGB (red, green, blue) color space to the HSV (hue, saturation, value) color space and use the linear function to stretch the original gray level dynamic range of the V component. Then, we estimate the illumination image via adaptive gamma correction and use the Retinex model to achieve the brightness enhancement. After that, we further stretch the gray level dynamic range to avoid low image contrast. Finally, we design another mapping function to achieve color saturation correction and convert the enhanced image from the HSV color space to the RGB color space after which we can obtain the clear image. The experimental results show that the enhanced images with the proposed method have better qualitative and quantitative evaluations and lower computational complexity than other state-of-the-art methods.

A G-Fresnel Optical Device and Image Processing Based Miniature Spectrometer for Mechanoluminescence Sensor Applications.

This paper presents a miniature spectrometer fabricated based on a G-Fresnel optical device (i.e., diffraction grating and Fresnel lens) and operated by an image-processing algorithm, with an emphasis on the color space conversion in the range of visible light. The miniature spectrometer will be cost-effective and consists of a compact G-Fresnel optical device, which diffuses mixed visible light into the spectral image and a µ-processor platform embedded with an image-processing algorithm. The RGB color space commonly used in the image signal from a complementary metal-oxide-semiconductor (CMOS)-type image sensor is converted into the HSV color space, which is one of the most common methods to express color as a numeric value using hue (H), saturation (S), and value (V) via the color space conversion algorithm. Because the HSV color space has the advantages of expressing not only the three primary colors of light as the H but also its intensity as the V, it was possible to obtain both the wavelength and intensity information of the visible light from its spectral image. This miniature spectrometer yielded nonlinear sensitivity of hue in terms of wavelength. In this study, we introduce the potential of the G-Fresnel optical device, which is a miniature spectrometer, and demonstrated by measurement of the mechanoluminescence (ML) spectrum as a proof of concept.

Integrating Image Analysis and Machine Learning for Moisture Prediction and Appearance Quality Evaluation: A Case Study of Kiwifruit Drying Pretreatment.

The appearance of dried fruit clearly influences the consumer's perception of the quality of the product but is a subtle and nuanced characteristic that is difficult to quantitatively measure, especially online. This paper describes a method that combines several simple strategies to assess a suitable surrogate for the elusive quality using imaging, combined with multivariate statistics and machine learning. With such a convenient tool, this study also shows how one can vary the pretreatments and drying conditions to optimize the resultant product quality. Specifically, an image batch processing method was developed to extract color (hue, saturation, and value) and morphological (area, perimeter, and compactness) features. The accuracy of this method was verified using data from a case study experiment on the pretreatment of hot-air-dried kiwifruit slices. Based on the extracted image features, partial least squares and random forest models were developed to satisfactorily predict the moisture ratio (MR) during drying process. The MR of kiwifruit slices during drying could be accurately predicted from changes in appearance without using any weighing device. This study also explored determining the optimal drying strategy based on appearance quality using principal component analysis. Optimal drying was achieved at 60 °C with 4 mm thick slices under ultrasonic pretreatment. For the 70 °C, 6 mm sample groups, citric acid showed decent performance.

Vehicle re-identification method based on multi-attribute dense linking network combined with distance control module.

Introduction: Vehicle re-identification is a crucial task in intelligent transportation systems, presenting enduring challenges. The primary challenge involves the inefficiency of vehicle re-identification, necessitating substantial time for recognition within extensive datasets. A secondary challenge arises from notable image variations of the same vehicle due to differing shooting angles, lighting conditions, and diverse camera equipment, leading to reduced accuracy. This paper aims to enhance vehicle re-identification performance by proficiently extracting color and category information using a multi-attribute dense connection network, complemented by a distance control module. Methods: We propose an integrated vehicle re-identification approach that combines a multi-attribute dense connection network with a distance control module. By merging a multi-attribute dense connection network that encompasses vehicle HSV color attributes and type attributes, we improve classification rates. The integration of the distance control module widens inter-class distances, diminishes intra-class distances, and boosts vehicle re-identification accuracy. Results: To validate the feasibility of our approach, we conducted experiments using multiple vehicle re-identification datasets. We measured various quantitative metrics, including accuracy, mean average precision, and rank-n. Experimental results indicate a significant enhancement in the performance of our method in vehicle re-identification tasks. Discussion: The findings of this study provide valuable insights into the application of multi-attribute neural networks and deep learning in the field of vehicle re-identification. By effectively extracting color information from the HSV color space and vehicle category information using a multi-attribute dense connection network, coupled with the utilization of a distance control module to process vehicle features, our approach demonstrates improved performance in vehicle re-identification tasks, contributing to the advancement of smart city systems.

Robust Rules for Optimal Colorimetric Sensing Based on Gold Nanoparticle Aggregation.

Spurred by outstanding optical properties, chemical stability, and facile bioconjugation, plasmonic metals have become the first-choice materials for optical signal transducers in biosensing. While the design rules for surface-based plasmonic sensors are well-established and commercialized, there is limited knowledge of the design of sensors based on nanoparticle aggregation. The reason is the lack of control over the interparticle distances, number of nanoparticles per cluster, or multiple mutual orientations during aggregation events, blurring the threshold between positive and negative readout. Here we identify the geometrical parameters (size, shape, and interparticle distance) that allow for maximizing the color difference upon nanoparticle clustering. Finding the optimal structural parameters will provide a fast and reliable means of readout, including unaided eye inspection or computer vision.

An Extension of Reversible Image Enhancement Processing for Saturation and Brightness Contrast.

This paper proposes a reversible image processing method for color images that can independently improve saturation and enhance brightness contrast. Image processing techniques have been popularly used to obtain desired images. The existing techniques generally do not consider reversibility. Recently, many reversible image processing methods have been widely researched. Most of the previous studies have investigated reversible contrast enhancement for grayscale images based on data hiding techniques. When these techniques are simply applied to color images, hue distortion occurs. Several efficient methods have been studied for color images, but they could not guarantee complete reversibility. We previously proposed a new method that reversibly controls not only the brightness contrast, but also saturation. However, this method cannot fully control them independently. To tackle this issue, we extend our previous work without losing its advantages. The proposed method uses the HSV cone model, while our previous method uses the HSV cylinder model. The experimental results demonstrate that our method flexibly controls saturation and brightness contrast reversibly and independently.

A new technique for stain-marking of seeds with safranine to track seed dispersal and seed bank dynamics.

Accurate tracking of seed dispersal is critical for understanding gene flow and seed bank dynamics, and for predicting population distributions and spread. Available seed-tracking techniques are limited due to environmental and safety issues or requirements for expensive and specialized equipment. Furthermore, few techniques can be applied to studies of water-dispersed seeds. Here we introduce a new seed-tracking method using safranine to stain seeds/fruits by immersing in (ex situ) or spraying with (in situ) staining solution. The hue difference value between pre- and post-stained seeds/fruits was compared using the HSV color model to assess the effect of staining. A total of 181 kinds of seeds/fruits out of 233 tested species of farmland weeds, invasive alien herbaceous plants and trees could be effectively stained magenta to red in hue (320-360°) from generally yellowish appearance (30-70°), in which the other 39 ineffectively-stained species were distinguishable by the naked eye from pre-stained seeds. The most effectively stained seeds/fruits were those with fluffy pericarps, episperm, or appendages. Safranine staining was not found to affect seed weight or germination ability regardless of whether seeds were stained ex situ or in situ. For 44 of 48 buried species, the magenta color of stained seeds clearly remained recognizable for more than 5 months after seeds were buried in soil. Tracking experiments using four species (Beckmannia syzigachne, Oryza sativa f. spontanea, Solidago Canadensis, and Acer buergerianum), representing two noxious agricultural weeds, an alien invasive plant, and a tree, respectively, showed that the safranine staining technique can be widely applied for studying plant seed dispersal. Identifying and counting the stained seeds/fruits can be executed by specially complied Python-based program, based on OpenCV library for image processing and Numpy for data handling. From the above results, we conclude that staining with safranine is a cheap, reliable, easily recognized, automatically counted, persistent, environmentally safe, and user-friendly tracking-seed method. This technique may be widely applied to staining most of the seed plant species and the study of seed dispersal in arable land and in disturbed and natural terrestrial or hydrophytic ecological systems.

Automatic segmentation of cervical region in colposcopic images using K-means.

Colposcopy is an important imaging modality for the detection of cervical lesions. The analysis of colposcopic images, especially the effective segmentation of the cervical region, has important clinical value in clinical application. A cervical segmentation method based on the HSV color mode is proposed, which can divide and extract the cervical region in the medical and anatomical sense. Firstly, the histogram threshold method is used to analyze the histogram (Y) of the colposcopic image. In order to achieve the removal of the mirror reflection pretreatment operation in the colposcopy image. Secondly, the Preprocessed RGB images is used. Then, the colposcopic image is converted into the HSV color space, and the V component is extracted using the K-means algorithm. Finally, using the area filter to smooth the edge, the segmented cervical region can be obtained. In our study, 110 standard colposcopy images, which were tagged by experts, were tested and verified. The segmentation results were analyzed and compared using dice coefficients, Jaccard coefficients, structural segmentation accuracy specificity, sensitivity, positive predictive value, and negative predictive value. Our experimental results show that the accuracy, specificity and sensitivity of the method are 87.25%, 81.99% and 96.70%, respectively. The effectiveness of the method in clinical segmentation was verified. Our study has demonstrated that cervical regional segmentation of colposcopic images based on HSV color space using K-means has high clinical utility and can help medical specialists in the diagnosis of cervical cancer.

Protein, enzyme and carbohydrate quantification using smartphone through colorimetric digitization technique.

In this paper the utilization of smartphone as a detection platform for colorimetric quantification of biological macromolecules has been demonstrated. Using V-channel of HSV color space, the quantification of BSA protein, catalase enzyme and carbohydrate (using D-glucose) have been successfully investigated. A custom designed android application has been developed for estimating the total concentration of biological macromolecules. The results have been compared with that of a standard spectrophotometer which is generally used for colorimetric quantification in laboratory settings by measuring its absorbance at a specific wavelength. The results obtained with the designed sensor is found to be similar when compared with the spectrophotometer data. The designed sensor is low cost, robust and we envision that it could promote diverse fields of bio-analytical investigations. Schematic illustration of the smartphone sensing mechanism for colorimetric analysis of biomolecular samples.

Detection of small colon bleeding in wireless capsule endoscopy videos.

In the recent years, wireless capsule endoscopy (WCE) technology has played a very important role in diagnosing diseases within the gastro intestinal (GI) tract of human beings. The WCE device captures images of the GI tract of patient with a certain frame rate. Physicians examine these images in order to find abnormalities in the GI tract. This examination process is very time consuming and hectic for the physician as a WCE device captures around 60,000 images on the average. At present, there are no standards defined for the WCE image classification. Computer aided methods help reducing the burden on the physicians by automatically detecting the abnormalities in the GI tract such as small colon bleeding. In this paper, a pixel based approach to detect bleeding regions in the WCE videos by using a support vector classifier is proposed. Threshold analysis in HSV color space is performed to compute the features for training an optimal support vector machine. The HSV features of the WCE images are fed to the trained support vector classifier for classification. Also, our method includes image enhancement and edge removal in WCE images, which is done prior to classification, for robust results. The method offers high sensitivity, specificity and accuracy in terms of correctly classifying images that contain bleeding regions as compared to another contemporary method. A detailed experimental analysis is also provided for the purpose of method evaluation.