Color Model Research Articles

Система для розпізнавання елементів одягу та їх кольорів на зображенні

The paper introduces a system for recognizing clothing items and their colors, developed using the Java programming language in the form of a web application, employing the Model-View-Controller (MVC) architecture. For development and project assembly convenience, the Gradle automatic build system was utilized. The system operates based on artificial neural network principles on the Clarifai platform, thoroughly examining the process of object detection in images and the system's real-world performance. InceptionV2 serves as the base model for image processing, incorporating the FPN technique to analyze images at different scales, resizing images up to 512 pixels, and training using stochastic gradient descent with hard negative mining. The system employs a median cut algorithm to determine the dominant color of clothing items. Determining the dominant color of each clothing item is done separately, with a focus on achieving 100 % recognition accuracy. An approach to determining the nearest color name based on the dominant color is implemented. This approach is based on calculating the Euclidean distance between two points in a three-dimensional space, iterating through 140 colors with RGB color model names to find the nearest color name. However, there are certain limitations in finding the nearest color name that result in a reduced accuracy of 60 %. The impact of various factors, such as lighting and image quality, is thoroughly examined in the context of their influence on the system's performance. The user interface is designed as an intuitive tool for interacting with the system, allowing users to check recognized clothing items and identified colors. Additionally, recommendations on basic color combinations that users can employ to enhance their clothing style are provided. The work includes an example of the system's application on a real image, visually demonstrating the results and describing the quality of clothing item and color recognition. One of the key features of the presented system is its flexibility and scalability. Recognition quality can be further improved through additional model training on larger datasets. The paper also addresses data processing and result analysis optimization. The novelty of the research results lies in the development of a comprehensive clothing item and color recognition system using artificial neural networks and the implementation of an approach to finding the nearest color with name. Completeness of the system refers to the entire image processing cycle within a single system, including clothing item location determination, clothing item classification, dominant color determination, and finding the nearest color with name for visual display to the user, along with providing basic color combinations for color correction or reassurance of color combinations' correctness. The practical significance of the research results lies in the development of the application's structure and algorithms, the software implementation of the system, which changes the approach to selecting and verifying color combinations. This approach provides users with clear color names, allowing them to create their own accurate color combinations for clothing sets or effectively search for clothing items online based on received colors. The system can also be used as a "data labeling" tool for preparing materials used in training machine learning models or neural networks. Future prospects include integration with other systems, expanding the image database to improve accuracy, and utilizing additional data sources to enhance the system's functionality.

CLASSIFICATION OF RICE USING CONVOLUTIONAL NEURAL NETWORK (CNN)

This paper describes the technique for automatic recognition and classification of different rice grain samples using neural network classifier. The Red Green Blue (RGB), Hue Saturation Intensity (HSI) and Hue Saturation Value (HSV) color models of the image were considered for extracting 18 color features. The classification was carried out using color and texture features separately. The color image was converted to Gray scale image and the Gray Level Co-occurrence Matrixes (GLCM) for four different directions was calculated. A total of eight texture features were calculated from the Co-occurrence matrices. Convolutional Neural Network (CNN) is used for the classification process. The classification accuracy with color features and texture features were compared. Result shows that the classification base on texture features outperform the color feature-based classification even with lesser number of features. It is found that Convolutional Neural network was able to classify two varieties of rice with 100% accuracy using texture features and the edge detection with Sobel and Canny edge detection of the fiber features in the food grain.

A social–ecological geography of southern Canadian lakes

Anthropogenic pressures, including urban and agricultural expansion, can negatively influence a lake's capacity to provide aquatic ecosystem services (ES). However, identifying lakes most at risk of losing their ES (i.e., higher vulnerability) requires integrating information on lake ecological state, global change threats, and ES use. Here, we provide a social–ecological framework that combines these features within a regional context by evaluating the ecological state of 659 lakes across Canada. Using the deviation of impacted lakes from reference ones, we identified much higher total nitrogen and chloride concentrations as the main indicators of an altered lake ecological state in all regions identified. Lake ecological state was mapped using an additive colour model along with regional scores of threat levels and recreational ES use. Urban and agriculturally developed areas were linked to higher lake vulnerability and ES loss. Lakes in Southern Ontario were most concerning, being highly altered, under threat, and heavily used. Lakes near coastal urban centers were altered and used, but less threatened, whereas those in the Prairies were altered and threatened, but less used. Our novel framework provides the first social–ecological geography of Canadian lakes, and is a promising tool to assess lake state and vulnerability at scales relevant for management.

CNN 기반 선박 형광 도막 두께 측정

To reduce the number of ship painting inspections in shipyards, there are trials to use visible fluorescent paint with a thickness of paint that can be visually inspected. However, due to the problem that the paint color varies depending on the illuminance and the type of light source, the reliability of the visual inspection is not consistent depending on the inspectors. Therefore, this study proposes a painting inspection method using machine learning technique instead of visual inspection. We propose automation of paint measurements using CNN model to find color variations in captured images according to the illuminance of paint. The actual thickness value of the paint was obtained from the specimen using a contact thickness measuring device. The color model was used to create a deep learning model suitable for the thickness characteristics of the image data. As a result, the proposed CNN model can measure the thickness of the paint within ±20 μm.

Development of an information support methodology for quality assessment of the prepress process

A methodology for improving the information support of the prepress system by introducing new information blocks to prevent errors and predict the final result before the start of prepress operations is presented. Conceptual provisions for assessing the quality of the prepress process have been developed. These conceptual provisions have become the basis for solving the problem of systemic and random errors at the planning stage, created conditions for a significant reduction in prepress time. This parameterized form allows a formal description of operations, the determination of relevant quality indicators based on the formed parameters, and the development of databases and knowledge of the information support system for the anticipatory quality control of prepress. Classification of factors that affect the quality of the prepress process was carried out. As a result, it is possible to exclude from further consideration the factors that are not important for assessing the prepress quality. The proposed factors form the basis of the knowledge base of the information system for assessing the quality of the prepress process. The method of “anticipatory” quality control was developed. This method provides the detection of “bottlenecks” in each technological operation of the prepress process, as a result of which random and system errors can be avoided. The resulting indicators of the proposed method are the probability of an error, the probability of error-free execution of the technological operation, the cost of controlling the technological operation, and the cost of eliminating the error. The information support method for the quality assessment of the prepress process is proposed. Implementation of the developed information support methodology for assessing the quality of the prepress process in the form of an information system for assessing the prepress quality was carried out. This information system allows automating the method of “anticipatory” quality control proposed in this study. Thus, on the example of the technological operation of color correction and color separation by the “CMYK color model” parameter based on the use of the expert knowledge base subsystem, the probability of an error is 1, and the cost of eliminating the error is $ 6.

Texture segmentation method for computer-assisted dermatologic diagnostic system

The wide spread of dermatological diseases is an important medical and social problem. Doctors note the constant growth of psoriasis among people of all ages. The psoriasis disease symptoms are similar to the symptoms of such diseases as eczema, atopic dermatitis and medicinal disease. Therefore, there is a high probability of an error in the disease diagnosis, which prevents the full treatment and prevention of the disease. Dermatological diagnostic systems are decision-making support systems for dermatologists when establishing a diagnosis and assessing the severity of the disease course. The development of new image processing methods for dermatological diagnostic systems is an important task, which allow to increase the accuracy of the diagnostic decision. In this work, the segmentation method of psoriasis images for systems of medical dermatological diagnostics based on a vector-difference approach to improve the quality of segmentation was developed. The vector-difference approach allows to calculate a certain texture feature of the image as a vector transformation of various texture features by linear algebra methods. Psoriasis disease images can be described by texture (spectral, statistical, spectral-statistical) and color, so it is proposed to take into account textural and color characteristics of images for image segmentation. The color models that are most often used in segmentation methods of psoriasis disease images were analyzed. Based on the analysis, the Hue-Saturation-Intensity color model was chosen.It is proposed to use spectral, statistical and spectral-statistical texture models and color characteristics of image pixels to represent psoriasis disease images. The developed segmentation method includes the following stages: image pre-processing; identification; vector-difference transformation; threshold processing. At the pre-processing stage, homomorphic filtering was applied to psoriasis disease images. The result of the identification stage is a set of features calculated by the textural and color characteristics for image objects. The vector-difference transformation converts the texture image into intensity. Threshold processing is performed with a global threshold. Experimental research of the proposed segmentation method of psoriasis disease images was performed. As a result of the experimental research, it was found that the probability of correct identification of psoriasis disease area on average is 0.97, the probability of a false alarm is about 0.05.

Eggplant Disease Detection Using Yolo Algorithm Telegram Notified

Eggplant (Solanum Melongena L) is a type of seasonal vegetable. Eggplants vary in shape and color from white to green to purple. Facts prove that eggplant plants face several threats to their survival and growth, including disease disturbances such as leaf beetles, fruit rot, mealybugs, etc. One way to treat and monitor eggplant plants is to install a camera to capture accurate and fast image data. The image captured by the camera will be digitally processed, which can be used by deep learning methods on object recognition and evaluation system in eggplant plants. Digital image processing can present the color accuracy of leaves, and fruit stems in eggplant plants for solutions and innovations. Digital image processing is the study of image processing techniques. Processed images are still images or images captured with a model camera. In this study, the images taken were of colored leaves attacked by eggplant rot and brown stem rot. The color model used in this study to transform images to color uses the red-green-blue color function. This study uses the YOLOv4 algorithm implemented on a mobile computing device such as Raspberry Pi 4 to select images of eggplant plants captured by a camera. The image data that has been processed will produce results in plant disease diagnoses based on the image captured by the camera. The results will be announced immediately or announced on the Telegram application. We can then detect diseases in eggplant plants and provide immediate preventative measures for diseased plants to maximize yield.

Application of regression and artificial neural network analysis of Red-Green-Blue image components in prediction of chlorophyll content in microalgae

This study presented a novel methodology to predict microalgae chlorophyll content from colour models using linear regression and artificial neural network. The analysis was performed using SPSS software. Type of extractant solvents and image indexes were used as the input data for the artificial neural network calculation. The findings revealed that the regression model was highly significant, with high R2 of 0.58 and RSME of 3.16, making it a useful tool for predicting the chlorophyll concentration. Simultaneously, artificial neural network model with R2 of 0.66 and low RMSE of 2.36 proved to be more accurate than regression model. The model which fitted to the experimental data indicated that acetone was a suitable extraction solvent. In comparison to the cyan-magenta-yellow-black model in image analysis, the red–greenblue model offered a better correlation. In short, the estimation of chlorophyll concentration using prediction models are rapid, more efficient, and less expensive.

Fire Detection using Image Processing

To reduce loss of life and property from fire, an early warning is an imperative. A fire detection system based on light detection and analysis is proposed in the paper. This system uses HSV and YCbCr color models with given conditions to separate orange, yellow, and high brightness light from background and ambient light. Fire growth is analysed and calculated based on frame differences. The overall accuracy from the experiments has been greater than 90%.

On the homotopy theory of equivariant colored operads

We build model structures on the category of equivariant simplicial operads with weak equivalences determined by families of subgroups, in the context of operads with a varying set of colors (and building on the fixed color model structures in the prequel). In particular, by specifying to the family of graph subgroups (or, more generally, one of the indexing systems of Blumberg-Hill), we obtain model structures on the category of equivariant simplicial operads whose weak equivalences are determined by norm map data.

Резонансные акустические и колориметрические характеристики древесины из ретросооружений

The reserves of resonance wood in the forests of the planet are limited, and in many countries they are completely absent, since it is formed only under certain habitat conditions in some trees with genetically determined properties. Therefore, the investigation of correlation between acoustic and colorimetric parameters of coniferous wood retreads has scientific and practical importance for the development of non-destructive methods of resonance material identification in the elements of old structures to reveal additional sources of raw materials. The article presents the results of complex dendro-acoustic and colorimetric studies of spruce and pine wood. The RGB levels for seasoned wood in the additive color model were found to be on average 12 % lower than those for recently harvested wood. The analysis showed that the interrelation of RGB levels with acoustic constant (K) is weak (r2 < 0.3) for all groups of seasoned wood samples. It was also found out that 39 % of the spruce samples and 23 % of the pine sapwood samples selected from the seasoned wood can be attributed to resonance wood, since their K ≥ 12 m4/kgs. Thus, old wooden structures made of spruce wood and pine sapwood to be slated for destruction due to the long term operation are of great practical interest as a source for producing resonance wood with unique dendro-acoustic properties. It was found that within the same wood species or part of a tree (sapwood, heartwood), the color remains constant, and only the tone varies. Although the results of the study of wood colorimetric properties and acoustic constant did not show a high level of correlation, the research in this area is still relevant. This result may be due to the imperfection of the color determination technique, namely the need to consider the different number and parameters of the annual rings that fall into the colorimeter measuring window. Overall, the results presented in this paper can be used in the development of new methods of express wood diagnostics in construction. For citation: Fedyukov V.I., Chernov V.Yu., Chernova M.S., Tsoy O.V. Resonance Acoustic and Colorimetric Characteristics of Wood in Old Structures. Lesnoy Zhurnal = Russian Forestry Journal, 2022, no. 6, pp. 164–177. (In Russ.). https://doi.org/10.37482/0536-1036-2022-6-164-177

An Adaptive Vector-Based Vehicles Detection for Urban Intersection Camera Sensors Under Nighttime Illumination

Vehicles detection via surveillance sensors is an intractable and essential task under nighttime urban intersection scenes. To efficiently resolve the problem that most of the current color or texture feature-based vehicles detection models easily suffer contamination from dynamic and sudden or gradual illumination variations on nighttime, an adaptive vector-based background model without complex artificial texture feature is proposed for vehicles detection in night intersection surveillance scenes. For each pixel of image, a series of filters are employed to take the values from the past at the same location and its neighborhood, and a vector is assigned to each filter to represent the neighborhood region information of the pixel. Then, the series of vectors comprising the background model and the vector of the current pixel is compared to the vectors of the background model based on the theorem of linear dependence to describe whether that pixel belongs to foreground or background. Eventually, to adapt the dynamic scenes, a random update scheme is employed to update the model. As our experimental results demonstrated in real-world nighttime urban intersection surveillance scenarios, the proposed model attains superior vehicles detection performance compared to other state-of-the-art algorithms based on extensive qualitative and quantitative evaluations.

Into OZ, color fusion MRI of the foot and ankle: A pictorial essay

“People would rather live in homes regardless of its grayness. There is no place like home.” (L. Frank Baum, The Wonderful Wizard of Oz) This article presents an explanation and introduction to use and benefits of Color Fusion MRI of the Foot and Ankle. There are conditions/abnormalities within the complex overlapping structure of the foot and ankle that involve more than one region or tissue type that if images are combined and different colors utilized are much easier to visualize and are a great aid in figuring out the challenging diagnostic puzzle. An efficient method for fusion anatomical and functional images is by constructing the fused image through the combination of parallel images in a multi-scale domain. The existing image fusion method combined with the image enhancement is used to obtain a final fused image with high-contrast detail information, simultaneously. However, the selection of color models and the formulation of transform based image fusion method plays a major role in the quality of fused images.

Early warning of Diaporthe infection in kiwifruit soft rot by plasmonic dimer-enhanced Raman spectroscopy.

Diaporthe genus is the dominant pathogens of kiwifruit soft rot with long incubation period and rapid onset and very hard to detect in advance. It is of great significance to develop point-of-care tests for disease prevention and field management. Here we screen a pair of genus-level primers for constructing plasmonic dimer Rayleigh/Raman spectroscopy, which enables rapid, specific, and sensitive detection of Diaporthe genus in real kiwifruits. Dark-filed Rayleigh scattering clearly visualizes the target-induced dimer assembly of nanoprobes. Plasmonic dimer-enhanced Raman spectroscopy has high specificity for 12 common isolates of Diaporthe genus in one batch samples with highly accurate results and much higher sensitivity than polymerase chain reaction (PCR). It realizes the recognition of Diaporthe infection at an early stage of 24h when the kiwifruits do not have any noticeable symptoms. It demonstrates a bright prospect of point-of-care tests for early warning of kiwifruit soft rot infection and quality control.

Infrared drying of apple (Malus domestica) slices: Effect on drying and color kinetics, texture, rehydration, and microstructure

AbstractIn this study, apple slices were dried using infrared (IR) and hot air techniques and comprehensively analyzed in terms of drying and product characteristics. The influence of IR power level (450–650 W) and hot air temperature (60–75°C) on mass transfer, color kinetics, product texture, microstructure, and rehydration characteristics was studied. The results indicated that drying time, color change, and energy requirement were lower in IR drying than in hot air drying. Moisture diffusivity was observed to increase with IR power (3.367 × 10−9–5.579 × 10−9 m2/s) and hot air temperature (1.288 × 10−9–2.387 × 10−9 m2/s). The activation energies of apple slices in IR and hot air drying were 11.94 and 21.90 kJ/mol, respectively. IR‐dried apple slices were more crispy, with a porous structure and higher rehydration ability. Experimental data were fitted to nine different thin‐layer drying and four‐color kinetic models using nonlinear regression analysis. The results of regression analysis indicated that the Midilli–Kucuk model is the best model to describe the drying behavior in both techniques. The color characteristics (L, a, and b) can be best explained by the modified color model and total color change by fraction conversion model for both IR and hot air drying of apple slices. This study revealed that IR drying of apple slices results in a better quality product in less time and energy as compared to hot air drying.Practical applicationsDrying is a vital food processing and preservation technique based on the principle of reducing the water content of the product. Although several drying techniques are available, there have been continuous efforts to improve drying methods in terms of energy efficiency and product quality attributes. The present work has been carried out considering the dearth of information on the influence of infrared power/intensity on the drying behavior and product quality of apple slices. Mass and color kinetics have been studied for a better understanding of the process, along with texture, microstructure, and rehydration properties. Our results showed that the process is superior, in terms of energy and product quality, as compared to other published work. It is concluded that infrared drying can be effectively used in the dehydration of apple slices on an industrial scale and can be promoted as a healthy alternative to fried snacks.

A Method of Detecting Candidate Regions and Flames Based on Deep Learning Using Color-Based Pre-Processing

Recently, object detection methods using deep learning have made significant progress in terms of accuracy and speed. However, the requirements of a system to provide real-time detection are somewhat high, and current methods are still insufficient to accurately detect important factors directly related to life and safety, such as fires. Therefore, this study attempted to improve the detection rate by supplementing the existing research to reduce the false detection rate of flame detection in fire and to reduce the number of candidate regions extracted in advance. To this end, pre-processing based on the HSV and YCbCr color models was performed to filter the flame area simply and strongly, and a selective search was used to detect a valid candidate region for the filtered image. In addition, for the detected candidate region, a deep learning-based convolutional neural network (CNN) was used to infer whether the object was a flame. As a result, it was found that the flame-detection accuracy of the model proposed in this study was 7% higher than that of the other models presented for comparison, and the recall rate was increased by 6%.

Late Deep Fusion of Color Spaces to Enhance Finger Photo Presentation Attack Detection in Smartphones

Finger photo recognition represents a promising touchless technology that offers portable and hygienic authentication solutions in smartphones, eliminating physical contact. Public spaces, such as banks and staff-less stores, benefit from contactless authentication considering the current public health sphere. The user captures the image of their own finger by using the camera integrated in a mobile device. Although recent research has pushed boundaries of finger photo matching, the security of this biometric methodology still represents a concern. Existing systems have been proven to be vulnerable to print attacks by presenting a color paper-printout in front of the camera and photo attacks that consist of displaying the original image in front of the capturing device. This paper aims to improve the performance of finger photo presentation attack detection (PAD) algorithms by investigating deep fusion strategies to combine deep representations obtained from different color spaces. In this work, spoofness is described by combining different color models. The proposed framework integrates multiple convolutional neural networks (CNNs), each trained using patches extracted from a specific color model and centered around minutiae points. Experiments were carried out on a publicly available database of spoofed finger photos obtained from the IIITD Smartphone Finger photo Database with spoof data, including printouts and various display attacks. The results show that deep fusion of the best color models improved the robustness of the PAD system and competed with the state-of-the-art.

A survey of recent advances in image steganography

AbstractOver the course of technological development, data security while transmission and storing has always been a field of continuous advancements and key importance. In the recent times, its importance has increased even further and the main reason behind this is the increase in the amount of data that is being transferred over internet and various networks. Although there are cryptographic approaches to handle these situations, steganography provides a different aspect to the solution. While cryptography encrypts the data such that the cipher text is visible to human eye, but the meaning of that encrypted text is never known. Steganography is the process of hiding data in plain sight inside some images, audio, video and so on, such that it is very difficult for the viewer to directly differentiate if there is some data hidden in the host or not. This paper consists a study of various methodologies to implement Image steganography like spatial domain manipulation, encryption based and deep learning approaches. The paper focuses upon recent advances in all these approaches for image steganography and aims to help other researches in this domain to understand recent developments as well future advancements possible in the same field. Furthermore, this paper illustrates about how different image formats effect various image steganography methods and how the various image color models interact with image steganography. Later this article sheds some light on the domain of steganalysis and reports about the fundamental principles as well as various implementations of steganalysis systems in recent times.

Geographical features of Tulipa suaveolens Roth (Liliaceae, Magnoliophyta) distribution by flower color across its European range

The range of the polychromous Tulipa suaveolens Roth comprises almost the whole Ponto-Caspian Steppe, from the south-east of Ukraine to western Kazakhstan. High variation in flower color is a unique feature of this species, but features of its geographical distribution remain unclear. We studied T. suaveolens tepal color variation in 56 natural populations across the European range. Tepal colors were detected from digital images using the Lab color model with two chromatic components, a (red color intensity) and b (yellow color intensity). A conclusion was made that, throughout the European range, an obvious T. suaveolens flower color gradient is expressed in the direction from the south-west to the north-east, along which the red chromatic component intensity decreases. A similar gradient is observed when the height above mean sea level increases. The chromatic component a was shown to have a spatial autocorrelation and to depend on the long-term bioclimatic environmental parameters, temperature and precipitation. Thus, the identified geographical trends in the T. suaveolens flower color distribution across the studied part of its range are a consequence of natural selection caused by these two abiotic factors.

Machine Learning Based Approach for Automated Cervical Dysplasia Detection Using Multi-Resolution Transform Domain Features

Pattern detection and classification of cervical cell dysplasia can assist with diagnosis and treatment. This study aims to develop a computational model for real-world applications for cervical dysplasia that has the highest degree of accuracy and the lowest computation time. Initially, an ML framework is created, which has been trained and evaluated to classify dysplasia. Three different color models, three multi-resolution transform-based techniques for feature extraction (each with different filters), two feature representation schemes, and two well-known classification approaches are developed in conjunction to determine the optimal combination of “transform (filter) ⇒ color model ⇒ feature representation ⇒ classifier”. Extensive evaluations of two datasets, one is indigenous (own generated database) and the other is publicly available, demonstrated that the Non-subsampled Contourlet Transform (NSCT) feature-based classification performs well, it reveals that the combination “NSCT (pyrexc,pkva), YCbCr, MLP” gives most satisfactory framework with a classification accuracy of 98.02% (average) using the F1 feature set. Compared to two other approaches, our proposed model yields the most satisfying results, with an accuracy in the range of 98.00–99.50%.