Image In RGB Color Space Research Articles

Fusion of visible and fluorescence imaging through deep neural network for color value prediction of pelletized red peppers.

A non-destructive method for determining the color value of pelletized red peppers is crucial for pepper processing factories. This study aimed to investigate the potentiality of visible and fluorescence images for the determination of color value of pelletized red pepper. The imaging problem, caused by the cylindrical shape and irregular cross-sectional features of the pelletized red peppers, was reduced through the extraction of an approximate plane region. To integrate the information in the visible and fluorescence images, a baseline convolutional neural network (CNN) architecture was designed and low level, middle level, and high level fusion models (denoted as LL-CNN, ML-CNN, and HL-CNN, respectively) were developed upon the baseline CNN. The effects of input image size and color space were examined. According to the training result, CNN fusion models were developed using visible image in L*a*b* color space and fluorescence image in RGB color space using 56×56 input image size. Among the three types of CNN fusion models, the HL-CNN obtained the best performance, resulting in Rv 2 of 0.828 and RMSEV of 0.351. This study suggests that the fusion of visible and fluorescence image through CNN is a practical approach to save testing time and replace traditional destructive method. The low cost and compact structure of the imaging systems can maintain the commercial appeal of pepper industry.

Filament-necking localization method via combining improved PSO with rotated rectangle algorithm for safflower-picking robots

Safflower is one of the most important specialty economic crops in the world. Safflower blossoms are harvested continuously for 3–5 times. Untimely picking can affect the opening of filaments in the next crop, resulting in reduced filament production and serious economic losses. However, the safflower images collected by picking robots were affected by lighting conditions or complex backgrounds. The existing segmentation algorithms have the problem of insufficient segmentation or over-segmentation, because of the small safflower size and localization area. Therefore, combining improved particle swarm optimization (PSO) with a rotated rectangle algorithm based on the filament-necking localization method for safflower-picking robots was proposed. The R-component image in RGB color space was extracted as a preprocessing sample by combining the color features of safflower. The inertia weights in the PSO algorithm are improved to enhance the performance of the algorithm. An adaptive nonlinear function is introduced to search for the optimal threshold and initially segmented to obtain the binary image. Then, the barycenter and minimum outer rectangle of the contour were set up using the rotated rectangle algorithm based on the geometric features of the filaments. The circular region of interest (Cir-ROI) of filament-necking is determined. The Zhang-Suen refinement algorithm was used for skeleton extraction to design an algorithm for localizing the picking point of safflower filaments. To test safflower images collected in complex environments, the results of average processing time were 0.14 s, and the average relative target area error rate was 19.33 %. Moreover, the localization accuracy of the picking point was 89.75 %. The filament-necking localization method provides a theoretical basis and experimental data support for damage reduction and efficient harvesting of safflower filaments.

A Novel Approach to Quantifying the Failure Modes of Concrete-Epoxy Interface

The failure or debonding of CEIs (Concrete-Epoxy Interfaces) in Fiber-Reinforced Polymer concrete (FRP) systems occurs in one or a combination of three modes: CC (Cohesive failure in Concrete), CE (Cohesive failure in Epoxy), and IF (Interfacial Failure). These failure modes are usually identified, and their relationships are established by human intuition, which is prone to subjectivity. This study proposes a novel method based on image processing techniques to analyze CEI fracture surfaces and evaluate their failure modes. The failure modes of CEI fracture surfaces of specimens from a 3PB (Three-Point Bending) experiment were assessed using an HVS, CIE L*a*b*, YCbCr, or RGB color space image segmentation-based image processing technique on the preprocessed images of the CEI failure sides. A manual approach was adopted to validate the accuracy of the proposed method. Comparing the failure mode (CE) obtained using the manual and the proposed methodology, an RMSE (Root Means Square Error) of 0.19, 0.10, 0.23, and 0.26 was obtained for HVS, CIE L*a*b*, YCbCr, or RGB color space, respectively. The epoxy area selected with CIE L*a*b* color space produced the most accurate evaluation of the failure modes. This study provides an accurate method of quantifying the failure modes of CEI fracture surfaces. The methodology proposed in this study is recommended for forensic investigations to understand better the possible causes of failure in externally bounded fiber-reinforced polymers.

Python programming code for stellar photometry in astrophysics teaching on a cloud computing service

Nowadays, there is various software used for both education and astronomy research. For photometry, licensed software and high-performance computer operating systems are required, which is a fund limitation for some schools in Thailand. Thus, in this article, we develop and present the Demonstration Photometry Scripts for Astrophysics Teaching (DPSAT version 1.0). The program is designed to work on cloud computing services via internet browsers to avoid hardware and operation requirement pain points. The DPSAT is programming on flexible, low-cost, on-trend language, Python, and Jupyter Notebook online editor. In advance, our new code supports the home-use image or video file format, i.e., jpg, png, or mp4. Thus, it will be more accessible for teachers and students who do not have standard astronomical instruments. The DPSAT measures the stellar light intensity from the time-series still-images or video files from a smartphone or other digital devices. The code can extract video files into sequenced still images, then transform the RGB color space images into greyscale. The light intensity signal of selected pixels is counted with a simple aperture method in time series. It shows the results, for example, the mean signal, standard variation, measured signal as light intensity versus time, and image of light sources. This will be fruitful for low-cost and easily accessible teaching of astrophysics subjects.

Presentation attack detection based on two-stream vision transformers with self-attention fusion

Aiming at the performance degradation of the existing presentation attack detection methods due to the illumination variation, a two-stream vision transformers framework (TSViT) based on transfer learning in two complementary spaces is proposed in this paper. The face images of RGB color space and multi-scale retinex with color restoration (MSRCR) space are fed to TSViT to learn the distinguishing features of presentation attack detection. To effectively fuse features from two sources (RGB color space images and MSRCR images), a feature fusion method based on self-attention is built, which can effectively capture the complementarity of two features. Experiments and analysis on Oulu-NPU, CASIA-MFSD, and Replay-Attack databases show that it outperforms most existing methods in intra-database testing and achieves good generalization performance in cross-database testing.

CArDIS: A Swedish Historical Handwritten Character and Word Dataset

This paper introduces a new publicly available image-based Swedish historical handwritten character and word dataset named Character Arkiv Digital Sweden (CArDIS) (https://cardisdataset.github. io/CARDIS/). The samples in CArDIS are collected from 64,084 Swedish historical documents written by several anonymous priests between 1800 and 1900. The dataset contains 116,000 Swedish alphabet images in RGB color space with 29 classes, whereas the word dataset contains 30,000 image samples of ten popular Swedish names as well as 1,000 region names in Sweden. To examine the performance of different machine learning classifiers on CArDIS dataset, three different experiments are conducted. In the first experiment, classifiers such as Support Vector Machine (SVM), Artificial Neural Networks (ANN), k-Nearest Neighbor (k-NN), Convolutional Neural Network (CNN), Recurrent Neural Network (RNN), and Random Forest (RF) are trained on existing character datasets which are Extended Modified National Institute of Standards and Technology (EMNIST), IAM and CVL and tested on CArDIS dataset. In the second and third experiments, the same classifiers as well as two pre-trained VGG-16 and VGG-19 classifiers are trained and tested on CArDIS character and word datasets. The experiments show that the machine learning methods trained on existing handwritten character datasets struggle to recognize characters efficiently on the CArDIS dataset, proving that characters in the CArDIS contain unique features and characteristics. Moreover, in the last two experiments, the deep learning-based classifiers provide the best recognition rates.

Using Texture Analyses and Statistical Classification for Detection Plant Leaf Diseases

The proposed method is based on classifying 15 types of plant leaf disease. Hue saturation value was used to delete the background and the healthy areas to show only the affected area in the image. Texture analyses adopted in image features extractions from the R component &G component &B component and creating 3 components which are RG and RB and GB color of the RGB color space images of diseased leaves. Building image classifier using statistical method for classification.

Steganography on Digital Color Image Using Modulo Function and Pseudo-Random Number Generator

In this era, many people exchange data digitally either through the internet or other communication channels. However, data sent digitally can be seen by unauthorized people. Therefore, data security is essential. One of the data security techniques is steganography. This study employs steganography based on Pixel Value Modification (PVM) method using the modulo function. This technique will be implemented to insert a text message in digital images in RGB color space. Besides, we use the Pseudo-Random Number Generator with a secret key to enhance the inserted messages’ security. To measure the performance of the proposed method, testing is carried out by comparing the stego image to cover image based on three criteria, namely imperceptibility, fidelity, and recovery. PVM method using modulo function successfully hides text that the length is less than 200 characters into 255×255-pixel color images. Imperceptibility testing is done by distributing a questionnaire to six people randomly. The results showed that all respondents answered that each test image's stego image and cover image had no difference. At the same time, the results of the fidelity test show that the MSE value is close to zero, and the PSNR value is above 40 dB. Furthermore, recovery testing will check whether the extracted message is the same as the secret message inserted in the cover image. The results showed that all messages inserted were extracted correctly for recovery criteria if given the correct secret key.

A double-pass fundus reflection model for efficient single retinal image enhancement

This study introduces a novel image formation model - the double pass fundus reflection (DPFR) model for retinal image enhancement (restoration). The DPFR model reveals the specific double pass fundus reflection feature that was hitherto neglected in modeling the light propagation of fundus imaging in all published reports on retinal image enhancement. Based on the DPFR model, the procedures of the proposed retinal image restoration algorithm are given. The failure of the dark channel prior on retinal images in RGB color space is clarified. While a solution about how to bypass the challenge is proposed. Each step of DPFR is tested experimentally with retinal images of different degraded situations to validate its robustness. Moreover, the DPFR method is tested on 906 images from five public databases. Six image quality matrixes including image definition, image sharpness, image local contrast, image multiscale contrast, image entropy, and fog density are used for objective assessments. The results are compared to the state-of-art methods, showing the superiority of DPFR over the others in terms of restoration quality and implementation efficiency. The MATLAB code is avaliable on https://github.com/ShuheZhang-MUMC/Double-Pass-Fundus-Reflection-model.

Real-time Hand Gesture Extraction Using Python Programming Language Facilities

Hand gesture recognition is one of communication in which used bodily behavior to transmit several messages. This paper aims to detect hand gestures with the mobile device camera and create a customize dataset that used in deep learning model training to recognize hand gestures. The real-time approach was used for all these objectives: the first step is hand area detection; the second step is hand area storing in a dataset form to use in the future for model training. A framework for human contact was put in place by studying pictures recorded by the camera. It was converted the RGB color space image to the greyscale, the blurring method is used for object noise removing efficaciously. To highlight the edges and curves of the hand, the thresholding method is used. And subtraction of complex background is applied to detect moving objects from a static camera. The objectives of the paper were reliable and favorable which helps deaf and dumb people interact with the environment through the sign language fully approved to extract hand movements. Python language as a programming manner to discover hand gestures. This work has an efficient hand gesture detection process to address the problem of framing from real-time video.

Underwater image enhancement using contrast limited adaptive histogram equalization and layered difference representation

Underwater images, which have low contrast and visibility as a result of selective attenuation based on the wavelength of the light passing through water, needs some corrections to extract meaningful information from them. In this paper, we aim to combine two different approaches; global and local contrast enhancement techniques, to obtain better visual quality while enhancing image contrasts on underwater images. While global technique (LDR) ensures the overall enhancement of the image, local technique (CLAHE) considers local brightness features of the image in RGB color space. The proposed method also applies local color correction on underwater image. While methods in the literature apply various approaches on the global histogram of channels, our method divides underwater image into non-overlapping sub-blocks and apply histogram equalization on them. The method uses HSV color space and especially S, V components for color correction. The results of the qualitative analysis show that it produces very good images, in contrast, color, and detail compared to other enhancement methods. The proposed method also decreases the effect of under- and over-enhanced areas and the blue-green effect on the output image. However, the visibility of the objects in the images are increased by color correction. For quantitative analysis, the proposed method produces the highest average value of entropy (7.83), EMEE (32.06), EME (40.97), average gradient (152.55), and Sobel count (130393) for 200 underwater images.

A Detection System for Rail Defects Based on Machine Vision

In this work, a machine vision based detection system is designed and applied to identify the region of rail track need to be polished. The CCD industrial camera is employed to obtain the RGB color space images on the railway line in real time. To reduce the effect of environmental brightness, the obtained RGB space images are transformed into HSI color space. Then, the noise was removed from the grayscale image by the Gaussian blur to remove the interference elimination of the acquisition signal. According to the calculated saturation and hue values, the threshold of saturation and hue values is set for the following image segmentation, respectively. At last, the accurate position information of the rail track region needs to be polished is obtained for the subsequent grinding processing. By employing the machine vision system, high-efficient automatic rail defects grinding could be realized.

Multiple macro nutrients deficiency identification using fuzzy C means

Every plant needs nutrients for their growth. The leaf analysis in a plant can present the need of a nutrient to be determined. The macronutrients deficiency in plants can be identified by specific type of color variations using leaves image. Early identification of macronutrients can help plants in becoming suitable for the harvest and lessen the utilization of farm inputs. The proposed system implements a novel approach to identify the lack of nutrients in maize leaf images. The maize leaf images are collected from Agricultural University, Coimbatore. The input images are preprocessed where the images in RGB color space are converted to Lab model. Then C means clustering is used to segment the nutrient deficient part from leaf image. The multi-Color space based feature Extraction is used to extract the features from deficient area of maize leaf image and type of nutrients deficiency is identified using Fuzzy rule-based system that gives the percentage of nutrient deficiency affected in maize image. Then, the performance of this work is evaluated using confusion matrix.

WITHDRAWN: Low light image contrast enhancement using advanced perfusion technique

Abstract The images captured tend to degrade due to the scattering and absorption of light. This may be due to the diminished colors, low light and undistinguishable objects in the image. For the improvement of quality of such degraded images, Advanced Perfusion technique is employed to enhance the contrast of the low light Image. In this technique, the input image in RGB color space is converted to LAB color spaces. Two versions of input image are considered, in which, the first image is processed for brightness contrast enhancement and the second image is processed for color contrast enhancement. For the enhancement of color contrast, the second image is passed through a Recursive Separated and Weighted Histogram Equalization, (RSWHE) followed by power law transformation. Two main features are extracted from both the images, namely, Laplacian weight map and Saliency weight map. These two features of the two LAB images are fused together to produce a final enhanced image, which has improved brightness contrast and color contrast. The proposed image enhancement technique is evaluated with various performance parameters such as Structural Similarity Index (SSIM), Peak Signal-to-Noise Ratio (PSNR), and Lightness Order Error (LOE). The performance of the proposed method is then compared with the existing methods for validation.

Intelligent Bio-Inspired Whale Optimization Algorithm for Color Image based Segmentation

Color image segmentation is widely used methods for searching of homogeneous regions to classify them into various groups. Clustering is one technique that is used for this purpose. Clustering algorithms have drawbacks such as the finding of optimum centers within a cluster and the trapping in local optima. Even though inspired meta-heuristic algorithms have been adopted to enhance the clustering performance, some algorithms still need improvements. Whale optimization algorithm (WOA) is recognized to be enough competition with common meta-heuristic algorithms, where it has an ability to obtain a global optimal solution and avoid local optima. In this paper, a new method for color image based segmentation is proposed based on using whale optimization algorithm in clustering. The proposed method is called the whale color image based segmentation (WhCIbS). It was used to divide the color image into a predefined number of clusters. The input image in RGB color space was converted into L*a*b color space. Comparison of the proposed WhCIbS method was performed with the wolf color image based segmentation, cuckoo color image based segmentation, bat color image based segmentation, and k-means color image based segmentation over four benchmark color images. Experimental results demonstrated that the proposed WhCIbS had higher value of PSNR and lower value of RMSR in most cases compared to other methods.

Empirical Examination of Color Spaces in Deep Convolution Networks

In this paper we present an empirical examination of deep convolution neural network (DCNN) performance in different color spaces for the classical problem of image recognition/classification. Most such deep learning architectures or networks are applied on RGB color space image data set, so our objective is to study DCNNs performance in other color spaces. We describe the design of our novel experiment and present results on whether deep learning networks for image recognition task is invariant to color spaces or not. In this study, we have analyzed the performance of 3 popular DCNNs (VGGNet, ResNet, GoogleNet) by providing input images in 5 different color spaces(RGB, normalized RGB, YCbCr, HSV , CIE-Lab) and compared performance in terms of test accuracy, test loss, and validation loss. All these combination of networks and color spaces are investigated on two datasets- CIFAR 10 and LINNAEUS 5. Our experimental results show that CNNs are variant to color spaces as different color spaces have different performance results for image classification task.

Visually meaningful encryption for color images by using Qi hyper-chaotic system and singular value decomposition in YCbCr color space

In traditional image encryption algorithms, the plain-text image is encrypted into a noise-like image, which exhibits an existence of the encryption to an attacker. To solve the problem, the concept of visually meaningful image encryption (VMIE) is introduced, which can avoid human eyes' detection and improve the security of encryption. In this paper, we propose an improved VMIE scheme for color images. In the present scheme, the original image in RGB color space is transformed into YCbCr color space. Discrete wavelet transform (DWT) and discrete cosine transform (DCT) are used in Cb and Cr regions. Singular value decomposition (SVD) is performed on each DCT block and the singular value is embedded into the cover image. Qi hyper-chaotic mapping is used in the pre-encryption to enhance the security. Experimental results show that the proposed scheme has better security and performance than existing VMIE schemes and can also effectively resist common attacks.

An Approach to Detect Image Forgery by Discrete Wavelet Decomposition

Image forgery or manipulation by using the multimedia technology is becoming a challenging issue. The most common type of image forgery is copy-move forgery where some part of one image is copied and spliced in the other image. In this article, first the images in RGB color space is converted into YCbCr color space and the four-level discrete wavelet transform (DWT) is implemented to detect image forgery. The output of the DWT is further processed by using the image gradient technique for the edge detection of spliced objects. Morphological operation and Wiener filtering are applied for locating the tempered region in the forged image. Sensitivity, specificity and accuracy calculated for spliced images of CASIA datasets are obtained 89%, 86% and 88% respectively.

The use of an extended set of key texture features Haralick in the diagnosis of plant diseases on leaf images

Proposed method based on fuzzy logic, including the operation of calculating the 36 key parameters of the R, G, B, RG, RB, GB color components of the original RGB color space images of plant leaves and the GLCM adjacency matrix, forming fuzzy conclusions about the type of plant disease, defusing using threshold binarization and majority voting on 6 parameters.

Farmland Extraction from High Spatial Resolution Remote Sensing Images Based on Stratified Scale Pre-Estimation

Extracting farmland from high spatial resolution remote sensing images is a basic task for agricultural information management. According to Tobler’s first law of geography, closer objects have a stronger relation. Meanwhile, due to the scale effect, there are differences on both spatial and attribute scales among different kinds of objects. Thus, it is not appropriate to segment images with unique or fixed parameters for different kinds of objects. In view of this, this paper presents a stratified object-based farmland extraction method, which includes two key processes: one is image region division on a rough scale and the other is scale parameter pre-estimation within local regions. Firstly, the image in RGB color space is converted into HSV color space, and then the texture features of the hue layer are calculated using the grey level co-occurrence matrix method. Thus, the whole image can be divided into different regions based on the texture features, such as the mean and homogeneity. Secondly, within local regions, the optimal spatial scale segmentation parameter was pre-estimated by average local variance and its first-order and second-order rate of change. The optimal attribute scale segmentation parameter can be estimated based on the histogram of local variance. Through stratified regionalization and local segmentation parameters estimation, fine farmland segmentation can be achieved. GF-2 and Quickbird images were used in this paper, and mean-shift and multi-resolution segmentation algorithms were applied as examples to verify the validity of the proposed method. The experimental results have shown that the stratified processing method can release under-segmentation and over-segmentation phenomena to a certain extent, which ultimately benefits the accurate farmland information extraction.