Grayscale Channel Research Articles

Classification of sugar using digital imaging and pattern recognition techniques

This work presents a new analytical methodology based on the use of digital images and multivariate analysis for the classification of sugar samples based on their color. Two studies were evaluated. The first involved the discrimination of refined (RS) and unrefined sugar (US) samples. The second study involved the classification of sugar samples into three different classes: refined sugar for the external market (EMRS); refined sugar for retail (RRS); and US. The frequency distribution of color indices in the red, green, blue, hue, saturation, intensity, and grayscale channels were obtained from digital images. Classification models based on partial least squares discriminant analysis (PLS-DA) and linear discriminant analysis (LDA) were developed. The best results obtained for the studies with two (97.4 % accuracy) and three (95.7 % accuracy) classes were achieved by applying the genetic algorithm (GA) in association with the linear discriminant analysis (GA-LDA) on the datasets. The proposed method offers several advantages, including low cost, simplicity, absence of chemical reagents, and non-destructiveness of the sample.

Secure Food-Allergen Determination by Combining Smartphone-Based Raw Image Analyses and Liquid Chromatography-Mass Spectrometry for the Quantification of Proteins Contained in Lateral Flow Assays.

The current food safety testing system, based on laboratory-based quantification, is difficult to scale up in line with the growth in the export market and does not enable traceability through the nodes of the food supply system. Screening assays, for example, lateral flow assays (LFAs), can improve traceability but often lack the required reliability to guarantee compliance. Here, we present an alternative pipeline for secure on-site compliance testing, using allergens as a case study. The pipeline features smartphone-driven LFA quantification and an liquid chromatography-mass spectrometry (LC-MS) method enabling direct quantification of the allergens contained in the LFA. The system enables swift and objective screening and provides a control measure to verify LFA assay reliability. For the smartphone assay, 8-bit RGB and grayscale colorimetric channels were compared with 16-bit raw intensity values. The latter outperformed RGB and grayscale channels in sensitivity, repeatability, and precision, while ratiometric ambient light correction resulted in excellent robustness for light-intensity variation. Calibration curves for peanut determination using two commercial LFAs featured excellent analytical parameters (R2 = 0.97-0.99; RSD 7-1%; LOD 3-7 ppm). Gluten determination with a third commercial LFA was equally established. A prediction error of 13 ± 11% was achieved for the best performing assay. Good performance-calibration curves (R2 = 0.93-0.99) and CVs (<15%)- were observed for the analyte quantification from the LFA by LC-MS. The LOD for the LC-MS assay was 0.5 ppm, well below the LODs reported for the LFAs. This method creates a digital, fast, and secure food safety compliance testing paradigm that can benefit the industry and consumer alike.

Fine‐grained semantic ethnic costume high‐resolution image colorization with conditional GAN

Grayscale image colorization, especially for ethnic costume images, is highly challenging due to its rich and complex color features. The existing image colorization methods usually take the costume image as a whole in practical applications that lead to the ignorance of the semantic information of different parts of the costume. It is known that each part's color distribution of the ethnic costume is different. So, the color mapping of other parts is also diverse, which is determined by distinctive ethnic characteristics. This study introduces fine-grained level semantic information and proposes a high-resolution image colorization model for ethnic costumes targeting enhancement, inspired by semantic-level colorization. The semantic information of different regions of ethnic costumes has a significant impact on the performance of the coloring task. Using Pix2PixHD as the backbone network, we create a new network architecture that maintains color distribution correspondence and spatial consistency of costume images using fine-grained semantic information. In our network, we take the splice result of fine-grained semantic for ethnic costume and grayscale image as the conditions and then feed them into the generative adversarial networks. We also discuss and analyze the influences of the grayscale channel and fine-grained semantics on discriminator. Extensive experiments demonstrate that our method performs well compared with other state-of-the-art automatic colorization methods.

Segmentasi Citra Kanker Serviks Menggunakan Markov Random Field dan Algoritma K-Means

Cervical cancer is a dangerous disease caused by malignant tumors that grow on the cervix and has globally attacked many women. Pap smear test is one of the early prevention efforts for cervical cancer. Medical personnel often have difficulty identifying images of cervical cancer cells. Several studies have used the K-Means clustering method to identify cervical cancer cell images from Herlev dataset. This study uses the Herlev dataset with the K-Means clustering algorithm and also used the Markov Random Field parameter as a feature for the process of identifying cervical cancer cell images. This study compared the results of the proposed method with some differences in the preprocessing process. The experimental results show an accuracy of 74,51% for RGB channels without low pass filter. Accuracy of 75,11% is obtained from the segmentation process using RGB channels with low pass filter. A further increase in accuracy was obtained by 75,76% when the segmentation process used the grayscale channel with low pass Filter. Based on the segmentation experiment with the highest segmentation accuracy results, the classification process using K-Nearest Neighbor (KNN) gives an accuracy of 89,29%.

Skewed distribution of leaf color RGB model and application of skewed parameters in leaf color description model

BackgroundImage processing techniques have been widely used in the analysis of leaf characteristics. Earlier techniques for processing digital RGB color images of plant leaves had several drawbacks, such as inadequate de-noising, and adopting normal-probability statistical estimation models which have few parameters and limited applicability.ResultsWe confirmed the skewness distribution characteristics of the red, green, blue and grayscale channels of the images of tobacco leaves. Twenty skewed-distribution parameters were computed including the mean, median, mode, skewness, and kurtosis. We used the mean parameter to establish a stepwise regression model that is similar to earlier models. Other models based on the median and the skewness parameters led to accurate RGB-based description and prediction, as well as better fitting of the SPAD value. More parameters improved the accuracy of RGB model description and prediction, and extended its application range. Indeed, the skewed-distribution parameters can describe changes of the leaf color depth and homogeneity.ConclusionsThe color histogram of the blade images follows a skewed distribution, whose parameters greatly enrich the RGB model and can describe changes in leaf color depth and homogeneity.

Low-energy proton calibration and energy-dependence linearization of EBT-XD radiochromic films.

In this work, we calibrate the newly developed EBT-XD radiochromic films (RCFs) manufactured by Gafchromictm using protons in the energy range of 4-10 MeV. Irradiation was performed on the 2 × 6 MV tandem linear accelerator located at the Université de Montréal. The RCFs were digitized using an Epson Perfection V700 flatbed scanner using both the red-green-blue and grayscale channels. The proton fluences were measured with Faraday cups calibrated in absolute terms. The linear energy transfer function within the active layer of the films was calculated using the mass stopping power tables coming from the PSTAR database from the National Institute of Standards and Technology (NIST) to allow retrieval of the deposited dose. We find that the calibration curves for 7 and 10 MeV protons are nearly equivalent. The 4 MeV calibration curves exhibit a quenching effect due to the Bragg peak that falls close to the active layer. A linearization of this energy dependence was developed using a semiempirical parametric model to allow the generation of calibration curves for any incident proton energy within the present range. Excellent correspondence (<5% dose difference for the same netOD) of the 10 MeV calibration curves was noted when compared to existing high-energy proton (148.2 MeV) calibration curves reported in the literature. Our calibration extends the range of operation of EBT-XD films to low-energy proton beam dosimetry.

Multifeature fusion for robust human detection in thermal infrared imagery

Despite all the significant advances in human detection in various environmental conditions, it is still a challenging task. Most of the human detection algorithms mainly use color information, which is not robust to lighting changes and varying colors under which such a detector should operate namely day and nighttime. This problem is further amplified with infrared (IR) imagery, which only contains grayscale information. The proposed algorithm for human detection uses intensity distribution, gradient, and texture features for effective detection of humans in IR imagery. For the detection of intensity, histogram information is obtained in the grayscale channel. For extracting gradients, we utilize the histogram of oriented gradients for better information in the various lighting scenarios. For extracting texture information, center-symmetric local binary pattern gives rotational invariance as well as lighting invariance for robust features under these conditions. Various binning strategies help keep the inherent structure embedded in the features, which provide enough information for robust detection of the humans in the scene. The features are then classified using an AdaBoost classifier to provide a tree-like structure for detection in multiple scales. The algorithm has been trained and tested on IR imagery and has been found to be fairly robust to viewpoint changes and lighting changes in dynamic backgrounds and visual scenes.

Quantification and identification of adulteration in the fat content of chicken hamburgers using digital images and chemometric tools

In this work, we developed an eco-friendly methodology for quantification and identification of adulteration in the fat content of chicken hamburgers by combining color histograms (in RGB, HSI, and Grayscale channels) obtained from digital images and chemometric tools. For this, 74 samples of chicken hamburgers with a fat content of 14.27–47.55% (w w−1) were studied, taking into account adulterations with a fat content higher than 20% (w w−1), as limited by Argentinean legislation. In both quantitative and qualitative approaches, chemometric models containing HSI histograms achieved the best results, because this is very suitable in situations where there is a need to separate the chromaticity from the intensity. In other words, the opacity of the sample surfaces increases with increasing fat content. PLS/HSI achieved the best quantification result with a R2 of 0.95, RMSEP of 2.01% w w−1, REP of 7.26% w w−1 and RPD of 4.47 in the prediction set, while SPA-LDA/Grayscale + HSI reached the most satisfactory in the test set with only one misclassified sample. Therefore, the proposed methodologies represent excellent alternatives to conventional Soxhlet extraction method, since they follow the primary principles of Green Analytical Chemistry, avoiding waste generation, besides not using either chemical reagents or solvents.

A Comparison of the YCBCR Color Space with Gray Scale for Face Recognition for Surveillance Applications

Face recognition is an important biometric method because of its potential applications in many fields, such as access control and surveillance. In this paper, the performance of the individual channels from the YCBCR colour space on face recognition for surveillance applications is investigated and compared with the performance of the grayscale. In addition, the performance of fusing two or more colour channels is also compared with that of the grayscale. Three cases with a different number of training images per persons were used as a test bed. It was found out that, the grayscale always outperforms the individual channel. However, the fusion of CBxCR with any other channel outperforms the grayscale when three images of the same class from the same database are used for training. Regardless of the cases used, the CBxCR channel always gave the best performance for the individual colour channels. It was found that, in general, increasing the number of fused channels increases the performance of the system. It was also found that the grayscale channel is the better choice for face recognition since it contains better quality edges and visual features which are essential for face recognition.

Utilizing text recognition for the defects extraction in sewers CCTV inspection videos

This paper proposed a novel automated framework for analyzing and tracking sewer inspection close-circuit television (CCTV) videos. The proposed model mainly supports the off-site examination and quality management process of the videos and enables efficient revaluation of CCTV videos to extract sewer condition data. The study discusses an important module for any automated analysis and defect detection in CCTV video. It includes two main modules: text recognition and cracks extraction. In the first module, multi-frame integration (MFI) was applied to reduce the background complexity, time and computational requirements needed for the video processing. Then maximally stable extremal regions (MSER) was used on the grayscale channel and HSV channel to effectively detect all the text edges. Saturation color channel was also applied to verify the detected text line and remove false alarms. In the second module, by utilizing the text information on each frame, the operator’s operation during the inspection is simulated which would indicate valuable clues about the location and severity of the cracks. The proposed methodology was validated using a set of video provided by the Korea Institute of Construction Technology.

A feature fusion strategy for human detection in omnidirectional camera imagery

Field of view of the traditional camera is limited such that usually more than three cameras is needed to cover the entire surveillance area. The use of multiple cameras usually requires more efforts regarding camera control and set up as well as they need additional algorithms to find the relationships among the images of different cameras. In this paper, we present a multi-feature algorithm that employs only one omnidirectional camera instead of using multiple cameras to cover the entire surveillance region. Here we use the image gradients, the local phase information based on phase congruency, the phase congruency magnitude, and the color features, and they are fused together to build one descriptor named as "Fused Phase, Gradients and Color features (FPGC). The image gradients, and local phase information based on phase congruency concept are used to extract the human body shape features. Either LUV or grayscale channel features are used according to the kind of camera used. The phase congruency magnitude and orientation of each pixel in the input image is computed with respect to its neighborhood. The resultant images are divided into local regions and the histogram of oriented phase, and the histogram of oriented gradient are determined for each local region and combined. A maximum pooling of the candidate features is generated for one channel of the phase congruency magnitude and the three LUV color channels. All these features are fed to a decision tree Adaboost classifier for training and classification between the classes. The proposed approach is evaluated on a challenging omnidirectional dataset and observed promising performance.

A Comparison of the YCBCR Color Space with Gray Scale for Face Recognition for Surveillance Applications

Face recognition is an important biometric method because of its potential applications in many fields, such as access control and surveillance. In this paper, the performance of the individual channels from the YCBCR color space on face recognition for surveillance applications is investigated and compared with the performance of the gray scale. In addition, the performance of fusing two or more color channels is also compared with that of the gray scale. Three cases with different number of training images per persons were used as a test bed. It was found out that, the gray scale always outperforms the individual channel. However, the fusion of CBxCR with any other channel outperforms the gray scale when three images of the same class from the same database are used for training. Regardless of the cases used, the CBxCR channel always gave the best performance for the individual color channels. It was found that, in general, increasing the number of fused channels increases the performance of the system. It was also found that the gray scale channel is the better choice for face recognition since it contain better quality edges and visual features which are essential for face recognition.

Absolute dosimetric characterization of Gafchromic EBT3 and HDv2 films using commercial flat-bed scanners and evaluation of the scanner response function variability.

Radiochromic films (RCF) are commonly used in dosimetry for a wide range of radiation sources (electrons, protons, and photons) for medical, industrial, and scientific applications. They are multi-layered, which includes plastic substrate layers and sensitive layers that incorporate a radiation-sensitive dye. Quantitative dose can be retrieved by digitizing the film, provided that a prior calibration exists. Here, to calibrate the newly developed EBT3 and HDv2 RCFs from Gafchromic™, we used the Stanford Medical LINAC to deposit in the films various doses of 10 MeV photons, and by scanning the films using three independent EPSON Precision 2450 scanners, three independent EPSON V750 scanners, and two independent EPSON 11000XL scanners. The films were scanned in separate RGB channels, as well as in black and white, and film orientation was varied. We found that the green channel of the RGB scan and the grayscale channel are in fact quite consistent over the different models of the scanner, although this comes at the cost of a reduction in sensitivity (by a factor ∼2.5 compared to the red channel). To allow any user to extend the absolute calibration reported here to any other scanner, we furthermore provide a calibration curve of the EPSON 2450 scanner based on absolutely calibrated, commercially available, optical density filters.

An analytical method for determination of quality parameters in cotton plumes by digital image and chemometrics

This paper proposes an analytical method based on the use of digital images and multivariate calibration for the determination of degree of yellowness (+b), reflectance (Rd) and content of wax (WAX) in samples of white cotton plumes and naturally colored plumes. Digital images acquisition of cotton plumes was carried out through a webcam and histograms containing frequency distributions of color index in red-green-blue (RGB), hue (H), saturation (S), value (V), and grayscale channels were obtained. Models of multivariate calibration based on Regression by partial least squares (PLS) using the complete histogram, as well as multiple linear regression (MLR) with the selection of variables by successive projections algorithm (SPA) were developed, validated and compared. Satisfactory prediction results were obtained for both models with Root Mean Square Error of Prediction (RMSEP) values varying from 0.76–0.83, 2.49–2.53% and 0.30–0.33% for +b, Rd and wax, respectively. According to a paired t-test at a 95% confidence level, no statistically significant difference was found between the predicted and reference values. An F-test at 95% confidence level does not indicate significant differences between the RMSEP values obtained with the full-histogram PLS and MLR-SPA models. It is a simple and low-cost method that does not use reagent, does not destroy the sample and performs analysis with comparable results to the reference method.

Digital image-based classification of biodiesel

This work proposes a simple, rapid, inexpensive, and non-destructive methodology based on digital images and pattern recognition techniques for classification of biodiesel according to oil type (cottonseed, sunflower, corn, or soybean). For this, differing color histograms in RGB (extracted from digital images), HSI, Grayscale channels, and their combinations were used as analytical information, which was then statistically evaluated using Soft Independent Modeling by Class Analogy (SIMCA), Partial Least Squares Discriminant Analysis (PLS-DA), and variable selection using the Successive Projections Algorithm associated with Linear Discriminant Analysis (SPA-LDA). Despite good performances by the SIMCA and PLS-DA classification models, SPA-LDA provided better results (up to 95% for all approaches) in terms of accuracy, sensitivity, and specificity for both the training and test sets. The variables selected Successive Projections Algorithm clearly contained the information necessary for biodiesel type classification. This is important since a product may exhibit different properties, depending on the feedstock used. Such variations directly influence the quality, and consequently the price. Moreover, intrinsic advantages such as quick analysis, requiring no reagents, and a noteworthy reduction (the avoidance of chemical characterization) of waste generation, all contribute towards the primary objective of green chemistry.

Classification of Individual Castor Seeds Using Digital Imaging and Multivariate Analysis

This paper presents a method based on digital imaging and multivariate analysis for the classification of castor seeds with respect to the cultivar type. For this purpose, two seed groups most commonly employed on Brazilian plantations were evaluated: BRS Nordestina and BRS Paraguacu cultivars (group I) and BRS Energia cultivar and CNPA 2009-7 genotype (group II). Images of these two different seed groups were recorded from a webcam and the frequency distribution of color indexes in the red-green-blue (RGB), hue (H), saturation (S), intensity (I), and grayscale channels were obtained. Pattern recognition methods based on partial least squares-discriminant analysis (PLS-DA) and linear discriminant analysis (LDA) were applied separately to each seed group. The best results were obtained by using the PLS-DA model, which correctly classified 97.5% and 98.8% of the prediction samples for groups I and II, respectively. The proposed method is simple, fast, non-destructive and non-expensive.

Classification of edible vegetable oil using digital image and pattern recognition techniques

This work presents a simple and non-expensive method based on digital image and pattern recognition techniques for the classification of edible vegetable oils with respect to the type (soybean, canola, sunflower and corn) and the conservation state (expired and non-expired shelf life). For this purpose, vegetable oil sample images were obtained from a webcam and the frequency distribution of color indexes in the red–green–blue (RGB), hue (H), saturation (S), intensity (I), and grayscale channels were obtained. Linear discriminant analysis (LDA) was employed in order to build classification models on the basis of a reduced subset of variables. For the purpose of variable selection, two techniques were utilized, namely the successive projection algorithm (SPA) and stepwise (SW) formulation. For the study evolving the classification with respect to oil type, LDA/SPA and LDA/SW models achieved a correct classification rate (CCR) of 95% and 90% respectively. For the identification of expired and non-expired samples, LDA/SPA models were found to be the best method for classifying sunflower, soybean and canola oils, achieving a CCR in the overall data set of 97%, 94% and 93%, respectively, while the LDA/SW correctly classified at 100% for corn oil data. These results suggest that the proposed method is a promising alternative for the inspection of authenticity and the conservation state of edible vegetable oils. As advantages, the method does not use reagents to carry out the analysis and laborious procedures for chemical characterization of the samples are not required.