RGB Digital Images Research Articles

Kiln-Net: A Gated Neural Network for Detection of Brick Kilns in South Asia

The availability of high-resolution satellite imagery has enabled several new applications such as identification of brick kilns for the elimination of modern-day slavery. This requires automated analysis of approximately 1 551 997 $\text{km}^2$ area within the “Brick-Kiln-Belt” of South Asia. Although modern machine learning techniques have achieved high accuracy for a wide variety of applications, problems involving large-scale analysis using high-resolution satellite imagery requires both accuracy as well as computational efficiency. We propose a coarse-to-fine strategy consisting of an inexpensive classifier and a detector, which work in tandem to achieve high accuracy at low computational cost. More specifically, we propose a two-stage gated neural network architecture called Kiln-Net . At the first stage, imagery is classified using the ResNet-152 model which filters out over $\text{99}\%$ of irrelevant data. At the second stage, a YOLOv3-based object detector is applied to find the precise location of each brick kiln in the candidate regions. The dataset, named Asia14 , consisting of $14\,000$ Digital Globe RGB images and 14 categories is also developed to train the proposed kiln-net architecture. Our proposed network architecture is evaluated on approximately $\text{3{,}300}$ km $^2$ region ( $337\,723$ image patches) from 14 different cities in five different countries of South Asia. It outperforms state-of-the-art methods employed for the recognition of brick kilns and achieved an accuracy of $\text{99.96}\%$ and average F1 score of 0.91. To the best of our knowledge, it is also $20\,$ x faster than existing methods.

Improved SERPENT Algorithm: Design to RGB Image Encryption Implementation

Gigantic data transfer over networks and internet turned in developing a large number of encryption techniques. As a consequence, in parallel, it gave birth to new plans of attacks. In facet of data security, the algorithms specifically Rijndael (Advanced Encryption Standard-AES), SERPENT and TWOFISH are equally considered to be the best. However, they pay less attention in some cryptographic areas like digital image encryption schemes because of their time consumption inferiority. In this paper, we introduce a time diminishing improved version of SERPENT algorithm depending upon chain ring-based substitution boxes (S-boxes) dealing with 8-bit vector instead of 4-bit. The multiplicative substructures in a chain ring have multiple generators and thus the chain ring based S-boxes improve the algebraic complexity of the cipher. Apart from it, the algorithm is utilized in a digital RGB image encryption application where throughout the chain ring operations are performed. The digital image analyses reveal that the proposed scheme consumes less time (i.e. 8.2 microseconds for encryption and 5.8 microseconds for decryption of a 128-bit block) as compared to the existing prominent RGB image encryption techniques. Furthermore, the investigation on the suggested RGB image encryption scheme it is observed that it has a great resistance against the statistical and the differential attacks.

Automated image processing framework for analysis of the density of fruiting bodies of Leptosphaeria maculans on oilseed rape stems

Understanding the transmission of plant pathogen inoculum during the periods when the host plants are not present is crucial for predicting the initiation of epidemics and optimizing mitigation strategies. However, inoculum production at the end of the cropping season, survival during the intercrop period, and the emergence or release of inoculum can be highly variable, difficult to assess, and generally inferred indirectly from symptom data. As a result, a lack of large datasets hampers the study of these epidemiological processes. Here, inoculum production was studied in Leptosphaeria maculans, the cause of phoma stem canker of oilseed rape. The fungus survives on stubble left in the field, from which ascospores are released at the beginning of the next cropping season. An image processing framework was developed to estimate the density of fruiting bodies produced on stem pieces following incubation in field conditions, and a quality assessment of the processing chain was performed. A total of 2540 standardized RGB digital images of stems were then analysed, collected from 27 oilseed rape fields in Brittany over four cropping seasons. Manual post‐processing removed 16% of the pictures, e.g. when moisture‐induced darkening of the oilseed rape stems caused overestimation of the area covered with fruiting bodies. The potential level of inoculum increased with increasing phoma stem canker severity at harvest, and depended on the source field and the cropping season. This work shows how image‐based phenotyping generates high‐throughput disease data, opening up the prospect of substantially increased precision in epidemiological studies.

Detection and quantification of adulterations in aged wine using RGB digital images combined with multivariate chemometric techniques

A method has been developed to authenticate aged high-quality wines and to quantify their potential adulterations through multivariate analysis and regression techniques applied to the obtained RGB digital images. Wines of pure Gran Reserva, Crianza, and Joven Rioja as well as synthetic adulterated Gran Reserva samples were studied. Digital images were obtained by a single and inexpensive lab-made device. Each sample was characterized by means of the 256 channels intensities from the RGB-colorgram. Multivariate image analysis revealed differences among the wine classes, and between genuine-aged and adulterated samples. Partial least squares regression was used to develop a model for estimating the adulteration degree of Gran Reserva wines. The model achieved good prediction (RMSEP = 1.6), appropriate precision (RSD = 2.5%) and suitable LOD (2.3%) to quantify cost-effective adulterations. The present method, due to simplicity and low cost, could provide an appropriate alternative to the traditional chemical authentication methods.

Low-cost assessment of grain yield in durum wheat using RGB images

The pattern of photosynthetic area of the canopy throughout the crop cycle is an important factor for determining grain yield in wheat. This work proposes the use of zenithal RGB images of the canopy taken in natural light conditions to derive vegetation indices as a low-cost approach to predict grain yield. A set of 23 varieties of durum wheat was monitored in three growing conditions (support irrigation, rainfed and late planting) and two sites (Aranjuez and Valladolid, Spain), totalling 6 field trials. For each plot, digital RGB images were taken periodically from seedling emergence to late grain filling. RGB-based Green Area (GA), Greener Area (GGA), Normalized Green Red Difference Index (NGRDI), Triangular Greenness Index (TGI) and a novel photosynthetic area index based on the CIE L*u*v* colour space (u*v*A) were compared to handheld spectroradiometer Normalised Difference Vegetation Index (NDVI) for reference. In the case of the irrigated and late planting trials the best phenotypic predictions of grain yield were achieved with the vegetation indices measured during the last part of the crop cycle (i.e. grain filling). For the rainfed trials the best phenotypic predictions were achieved with indices measured earlier (around heading). Among all the evaluated indices, the novel index performed the best. Considering the heritabilities of the evaluated RGB indices and their genetic correlations with grain yield, index-based predictions of grain yield were best in the early crop stages for both rainfed and irrigated conditions, while for late planting indices measured at different crop stages performed equally well.

Color Measurement of Segmented Printed Fabric Patterns in Lab Color Space from RGB Digital Images

Colors of textile materials are the first parameter of quality evaluated by consumers and a key component considered in selecting printed fabric. In the textiles industry, digital printed fabric analysis is one of the basic elements in successfully utilizing a color mechanism scheme and objectively evaluating fabric color alterations. Precise color measurement, however, is mostly used in sample analysis and quality inspection which help to produce reproducible or similar product. It is important that for quality inspection, the color of the product should be measured as a necessary requirement of quality control whether the product is to be accepted or not. Presented in this study is an unsupervised segmentation of printed fabrics patterns using mean shift algorithm and color measurements over the segmented regions of printed fabric patterns. The results established a consistent and reliable color measurement of multiple color patterns and appearance with the established range without any interactions.

Generalized Chaos Synchronization Circuit Simulation and Asymmetric Image Encryption

Generalized chaos systems have more complex dynamic behavior than conventional chaos systems. If a generalized response system can be synchronized with a conventional drive system, the flexible control parameters and unpredictable synchronization state will increase significantly. The study first constructs a four-dimensional nonlinear dynamic equation with quadratic variables as a drive system. The numerical simulation and analyses of the Lyapunov exponent show that it is also a chaotic system. Based on the generalized chaos synchronization (GCS) theory, a four-dimensional diffeomorphism function is designed, and the corresponding GCS response system is generated. Simultaneously, the structural and synchronous circuits of information interaction and control are constructed with Multisim™ software, with the circuit simulation resulting in a good agreement with the numerical calculations. In order to verify the practical effect of generalized synchronization, an RGB digital image secure communication scheme is proposed. We confuse a 24-bit true color image with the designed GCS system, extend the original image to 48-bits, analyze the scheme security from keyspace, key sensitivity and non-symmetric identity authentication, classical types of attacks, and statistical average from the histogram, image correlation. The research results show that this GCS system is simple and feasible, and the encryption algorithm is closely related to the confidential information, which can resist the differential attack. The scheme is suitable to be applied in network images or other multimedia safe communications.

A freshness indicator for monitoring chicken-breast spoilage using a Tyvek® sheet and RGB color analysis

Abstract In this study, a simple, inexpensive, and high-confidence freshness indicator was developed to detect chicken-breast spoilage using a pH-sensitive dye, bromocresol green (BCG); color changes in the freshness indicator were analyzed using RGB digital images. The indicator consisted of a three-layered structure, using a Tyvek® sheet with high gas and vapor permeability as the inner layer, BCG-immobilized with the binding polymer as the color-changing layer, and a low-density polyethylene film as the outer layer. Color changes in the indicator were correlated with CO2 and total volatile basic nitrogen contents and bacterial growth in the chicken breast, which were quantified using digital images captured on a smartphone. When the indicator turned yellow from green, it indicated spoilage; further, a large difference was observed in the RGB values. The proposed freshness indicator with no dye migration is highly feasible, safe, and reliable for the real-time monitoring of quality changes in chicken breasts. Digital image processing via a smartphone makes it a simple and accurate method, allowing consumers to judge food quality easily and reliably.

Double random phase encoding with a Poisson-multinomial distribution for efficient colorful image authentication

In this paper, a new integrated approach is proposed for authenticating digital color image. Here, a new technique, Poisson-Multinomial Distribution (PMD) is introduced for the first time in image processing. It is used for Photon Counting Imaging (PCI) which is integrated with Double Random Phase Encoding (DRPE) scheme in this system. The main goal of this proposal is to establish an image authentication architecture which will only work for 3D digital color images. It is a simplified way for applying PCI scheme on three channels of a 3D image simultaneously. The system will generate a stationary white noise as a final output, which will be difficult to decode for a third party attacker. This proposed scheme works directly with the original digital RGB image. At first, the system encrypts those three channels of the image individually with DRPE method without separating them and the amplitude part of the encrypted image is photon counted using PMD. Finally, to obtain an optimal result, a probability density function is used. On the authentication part, the reference digital image is encrypted by the same keys as the original image and then both of the encrypted images are compared with a statistical nonlinear correlation method. The numerical experiments say that, this proposed PMD based method is proven to be a good and simplified one that can be used to encrypt digital color images. In addition, even if the number of photons is really low, this new system can successfully differentiate between true class and false class images. To prove its efficiency we have also shown some experimental results under different situations.

Study of Cleaning Products Effect on Corrosion By Means of RGB Video-Electrochemical Techniques

A digital endoscope camera with a USB connection was used to capture full-colored digital videos during spontaneous and electrochemical forced corrosions of technological metals digital videos at suitable regular intervals using a free software. The set of images, which included the complete interval time of the experiences, were automatically read and converted into intensity arrays by Mathcad® v14 software. After selecting the area under study, cropped images were decomposed into red, green and blue components as intensity subarrays using the Image Processing Functions included in the software. Values of arrays were classified between 0 and 255 intensity levels. Electrochemical experiments were performed with a PAR 273A potentiostat/galvanostat. The electrochemical cell was a 2×2 cm high-transmittance glass cell (Hellma, OG quality). Digital endoscope camera (EPA-503278) with USB connection was used to capture 640×480 color digital videos up to 30 framessecond. Digital images obtained from video were jpg RGB 8 bits per channel. The electrochemical cell was illuminated in a home-made white box with a T5 6400K fluorescent tri-phosphor lamp (color rendering index 80). An electrode of reference Hg/HgSO4/ K2SO4 (saturated) was used [1],The results of this work shown as an example are the effect of some types of commercial industrial cleaning products that differ in the contents of aggressive substances such as peroxides, hypochlorite or chloride ion on three types of metallic materials widely used in industrial factory: aluminum, steels stainless steel and brass. The reference electrode Hg / HgSO4 / K2SO4 (saturated) was used. For comparative purposes, the quantification of the potential for zero intensity prior to the anodic dissolution provides an approximate value of the corrosion potential. The height of the anodic dissolution peak gives an approximate value of the apparent corrosion rate. Both measures in potentiodynamic regime involve a rapid evaluation of the corrosion kinetics of the metal in the test media. The intensity changes in red (R), green (G) and blue (B) give a quantitative idea of changes in surface composition or within the dissolution. When convenient, the experiment can focus on the potential range of the anodic oxidation signal, at a maximum rate of 1 millivolt per second, and then calculate the exchange current (corrosion), the equilibrium potential (corrosion potential) and the slope of Tafel by a classical way. Color histograms of RGB intensity subarrays showed the statistical distribution of Intensities-Red, green and blue component analysis from RGB digital image video opens up a new direction to evaluate and quantify the electrochemical corrosion of large electrode areas and to improve the inhibitors of corrosion and surface treatments fields. RGB digital image video is a very versatile and low cost optical technique for investigating the electrochromic and corrosion processes. This mixed electrochemical-optical methodology can be especially indicated for the characterization of large surface samples, even dark substrates [2]. [1] J. Agrisuelas, J. J. García-Jareño, F. Vicente Spatiotemporal colorimetry to reveal electrochemical kinetics of poly(o-toluidine) films along ITO surface. Electrochim. Acta, 269 (2018) 350-358[2] J. Agrisuelas, J.J. García-Jareño, E. Perianes, F. Vicente, Use of RGB digital video analysis to study electrochemical processes involving color changes, Electrochem. Commun. 78 (2017) 38–42. This work was supported by Betelgeux -IVACE IMDITA2016/155 and MINECO-FEDER CTQ2015-71794-R

Quantification of electrochromic kinetics by analysis of RGB digital video images

A new, rapid, and practical procedure to obtain quantitative kinetic information of electrochromic materials is proposed. When the transition between two colors on a colored electrode surface is monitorized by digital video, the analysis of pixel color intensity (I) in sequential RGB digital images allows us to precise the half-life time (t1/2) of the color change since a maximum value of the standard deviation of the pixel intensity (σ(I)) is reached. We simulated and analyzed the evolution of standard deviation during an electrochromic transition in different circumstances (kinetic and diffusion control). As an example of application, electrochromic transitions of a thin film model, the Prussian blue (PB), were digitally recorded during potential step chronoamperometry. From obtained t1/2, we calculated kinetic constants of the first order blue-to-transparent transition at different potential steps. The results suggest the great potential of this novel methodology to quantify the kinetics of electrochromic changes, even, in proof-of-concepts devices.

Emulsified jet fuel flame characterization with image processing

This work describes a RGB digital image processing approach of emulsified jet fuel flame, which allows the characterization of the combustion phenomenon in the case of new fuels through color chemiluminiscence measurements. By applying RGB techniques, the image processing of the flame reveals useful parameters in an effective and cost-efficient technique for the determination of relevant chemical species, such as CH* and C2*, equivalence ratio, and temperature estimation.Second generation emulsified aviation fuels containing water-jet fuel have been a challenge for simultaneous thrust augmentation and pollution diminution, with subsequent cost reduction and fossil fuel dependence. Testing new fuels would normally require expensive equipment and reliable investigation techniques, while image processing proved to be a reliable method for the estimation of combustion chemical species and temperature in the case of classic fuels.For the combustion behavior of emulsified jet fuel, a co-annular spray burner was used, allowing the complex investigation with UV-VIS spectrometer and flame photography. RGB image processing techniques showed good agreement with more complicated diagnosis tools, such as spectrometers.

Using image analysis for quantitative assessment of needle bladder rust disease of Norway spruce.

High elevation spruce forests of the European Alps are frequently infected by the needle rust Chrysomyxa rhododendri, a pathogen causing remarkable defoliation, reduced tree growth and limited rejuvenation. Exact quantification of the disease severity on different spatial scales is crucial for monitoring, management and resistance breeding activities. Based on the distinct yellow discolouration of attacked needles, it was investigated whether image analysis of digital photographs can be used to quantify disease severity and to improve phenotyping compared to conventional assessment in terms of time, effort and application range. The developed protocol for preprocessing and analysis of digital RGB images enabled identification of disease symptoms and healthy needle areas on images obtained in ground surveys (total number of analysed images n = 62) and by the use of a semiprofessional quadcopter (n = 13). Obtained disease severities correlated linearly with results obtained by manual counting of healthy and diseased needles for all approaches, including images of individual branches with natural background (R 2 = 0.87) and with black background (R 2 = 0.95), juvenile plants (R 2 = 0.94), and top views and side views of entire tree crowns of adult trees (R 2 = 0.98 and 0.88, respectively). Results underline that a well‐defined signal related to needle bladder rust symptoms of Norway spruce can be extracted from images recorded by standard digital cameras and using drones. The presented protocol enables precise and time‐efficient quantification of disease symptoms caused by C. rhododendri and provides several advantages compared to conventional assessment by manual counting or visual estimations.

A simple 2D composite image analysis technique for the crystal growth study of L-ascorbic acid.

This work was destined for 2D crystal growth studies of L-ascorbic acid using the composite image analysis technique. Growth experiments on the L-ascorbic acid crystals were carried out by standard (optical) microscopy, laser diffraction analysis, and composite image analysis. For image analysis, the growth of L-ascorbic acid crystals was captured as digital 2D RGB images, which were then processed to composite images. After processing, the crystal boundaries emerged as white lines against the black (cancelled) background. The crystal boundaries were well differentiated by peaks in the intensity graphs generated for the composite images. The lengths of crystal boundaries measured from the intensity graphs of composite images were in good agreement (correlation coefficient "r" = 0.99) with the lengths measured by standard microscopy. On the contrary, the lengths measured by laser diffraction were poorly correlated with both techniques. Therefore, the composite image analysis can replace the standard microscopy technique for the crystal growth studies of L-ascorbic acid.

Tuber borchii Vitt. mycorrhiza protects Cistus creticus L. from heavy metal toxicity

Heavy metals (HMs), such as copper, zinc, lead, mercury and cadmium, are the most abundant and dangerous inorganic environmental pollutants. Growing pieces of evidence suggest that mycorrhizal fungi can alleviate metal toxicity in plants. In this study, we focused attention on the ectomycorrhizal (ECM) fungus Tuber borchii Vitt., which is widespread in Italy and is of great ecological interest because of the mutualistic associations and the advantages it provides to host plants. Seedlings of the Mediterranean shrub Cistus creticus L., mycorrhized and non mycorrhized with the ECM fungus T. borchii, were treated with HMs (zinc, lead and chromium). HMs induced leaves’ chlorosis in non mycorrhized seedlings; while no significant difference was observed in pigmentation of mycorrhized seedlings’ leaves. This observation was confirmed by Euclidean Distance of color measurements in L*a*b* units from RGB digital images of leaves. The decrease in leaves pigmentation observed in HM treated non mycorrhized seedlings strongly correlated with a reduced expression of key genes associated with chlorophyll biosynthesis; instead, no significant variation of gene expression was detected in mycorrhized seedlings treated with HMs.Finally, a comet test on nuclei isolated from leaves of mycorrhized and non mycorrhized C. creticus seedlings, treated or not with HMs, confirmed that the DNA damage induced by HMs stress was lower in mycorrhized than in non mycorrhized seedlings. Taken together, the data suggest the involvement of the mycorrhizal fungus T. borchii in the improvement of HM stress tolerance in C. creticus host.

Evaluation of peripheral vasodilative indices in skin tissue of type 1 diabetic rats by use of RGB images

We investigated a method to evaluate the arterial inflow and the venous capacitance in the skin tissue of streptozotocin-induced type 1 diabetic rats from RGB digital color images. The arterial inflow and the venous capacitance in the dorsal reversed McFarlane skin flap are calculated based on the responses of change in the total blood concentration to occlusion of blood flow to and from the flap tissues at a pressure of 50 mmHg. The arterial inflow and the venous capacitance in the skin flap tissue were significantly reduced in type 1 diabetic rat group compared with the non-diabetic rat group. The results of the present study indicate the possibility of using the proposed method for evaluating the peripheral vascular dysfunctions in diabetes mellitus.

Non-Contact Measurement of Cereal Quality by Image Sensing and Numerical Regression Techniques

In this paper, digital image processing techniques are applied to measure some of the quality parameters of the durum wheat semolina. One of these parameters is the semolina colour value in the lab colour space L*a*b*, which is the commonly employed colour space in food field. Several numerical methods are developed and analysed for mapping the RGB digital images to L*a*b*. These methods are direct, polynomial regression, and neural network methods. The accuracy of each method is obtained with respect to the measured L*a*b* values captured with a Chroma-Meter instrument. The numerical models outcomes showed lowest colour deviations of 0.72. The results also demonstrated a significant effect of the training data set on the numerical L*a*b* outputs. Moreover, a partial least-squares regression model was developed to numerically predict the β–carotene content in semolina, as another important quality parameter. The model proved a correlation coefficient of 0.94 between numerical predictions and experimental measurements according to the ICC standard method 152 for extracting the durum carotenoids, thus bears a high potential for facilitating carotene detection in durum.

Measurement of food colour in L*a*b* units from RGB digital image using least squares support vector machine regression

The aim of this work is to evaluate the potential of least squares support vector machine (LS-SVM) regression to develop an efficient method to measure the colour of food materials in L*a*b* units by means of a computer vision systems (CVS). A laboratory CVS, based on colour digital camera (CDC), was implemented and three LS-SVM models were trained and validated, one for each output variables (L*, a*, and b*) required by this problem, using the RGB signals generated by the CDC as input variables to these models. The colour target-based approach was used to camera characterization and a standard reference target of 242 colour samples was acquired using the CVS and a colorimeter. This data set was split in two sets of equal sizes, for training and validating the LS-SVM models. An effective two-stage grid search process on the parameters space was performed in MATLAB to tune the regularization parameters γ and the kernel parameters σ<sup>2</sup> of the three LS-SVM models. A 3-8-3 multilayer feed-forward neural network (MFNN), according to the research conducted by León <em>et al.</em> (2006), was also trained in order to compare its performance with those of LS-SVM models. The LS-SVM models developed in this research have been shown better generalization capability then the MFNN, allowed to obtain high correlations between L*a*b* data acquired using the colorimeter and the corresponding data obtained by transformation of the RGB data acquired by the CVS. In particular, for the validation set, R<sup>2</sup> values equal to 0.9989, 0.9987, and 0.9994 for L*, a* and b* parameters were obtained. The root mean square error values were 0.6443, 0.3226, and 0.2702 for L*, a*, and b* respectively, and the average of colour differences ΔE<sub>ab</sub> was 0.8232±0.5033 units. Thus, LS-SVM regression seems to be a useful tool to measurement of food colour using a low cost CVS.

Security of RGB image data by affine hill cipher over [formula omitted] and [formula omitted] domains with Arnold transform

These days, security of sensitive information is a major concern. Cryptography allows transmission of data from one end to the other, both, securely and secretly. Here we have designed a novel cryptosystem for security of RGB images by affine hill cipher (AHC) over SLn(Fq) and Mn(Fq) domains with Arnold transform (AT). The security of color images is designed by the existing techniques with the help of keys only, but proposed cryptosystem provides security of RGB image data on the basis of the keys and the arrangement of AHC parameters. Additionally, key multiplication side (pre or post) with a RGB image data is also inevitable to know, to correctly decrypt the encrypted image data. In this approach, we have considered multiplicative keys of affine hill cipher from SLn(Fq) domain and additive keys of affine hill cipher from Mn(Fq) domain, which provides exorbitant key space for the proposed cryptosystem. Experimental results and security analysis of the cryptosystem are given to the justification for the stalwartness of the proposed cryptosystem. This method would have large potential usage in digital RGB image processing and security.

Application of algebra and discrete wavelet transform in two-dimensional data (RGB-images) security

In this cryptosystem, we have considered RGB images for two-dimensional (2D) data security. Security of RGB images during transmission is a major concern, discussed globally. This paper proposes a novel technique for color image security by random hill cipher (RHC) over SLn(𝔽) domain associated with 2D discrete wavelet transform. Existing techniques have discussed the security of image data on the basis of the keys only (which provide only one layer of security for image data), but in the proposed cryptosystem, the keys and the arrangement of RHC parameters are imperative for correct decryption of color image data. Additionally, key multiplication side (pre or post) with the RGB image data should inevitably be known, to correctly decrypt the encrypted image data. So, the proposed cryptosystem provides three layers of security for RGB image data. In this approach, we have considered keys from the special linear group over a field 𝔽, which provides enormous key-space for the proposed cryptosystem. A computer simulation on standard examples and results is given to support the fixture of the scheme. Security analysis, and detailed comparison between formerly developed techniques and proposed cryptosystem are also discussed for the robustness of the technique. This method will have large potential usage in the digital RGB image processing and the security of image data.