Sobel Operator Research Articles

Searching surveillance video contents using convolutional neural network

Manual video inspection, searching, and analyzing is exhausting and inefficient. This paper presents an intelligent system to search surveillance video contents using deep learning. The proposed system reduced the amount of work that is needed to perform video searching and improved the speed and accuracy. A pre-trained VGG-16 CNNs model is used for dataset training. In addition, key frames of videos were extracted in order to save space, reduce the amount of work, and reduce the execution time. The extracted key frames were processed using the sobel operator edge detector and the max-pooling in order to eliminate redundancy. This increases compaction and avoids similarities between extracted frames. A text file, that contains key frame index, time of occurrence, and the classification of the VGG-16 model, is produced. The text file enables humans to easily search for objects of interest. VIRAT and IVY LAB datasets were used in the experiments. In addition, 128 different classes were identified in the datasets. The classes represent important objects for surveillance systems. However, users can identify other classes and utilize the proposed methodology. Experiments and evaluation showed that the proposed system outperformed existing methods in an order of magnitude. The system achieved the best results in speed while providing a high accuracy in classification.

Application of Gradient-based Edge Detection on Defined Surface Feature Boundary

In surface topography, the detection of edges or surface feature boundary should be made prior to the characterization process to ensure precise measurement of the surface feature. However, problem arises when the existing edge detection from the image processing area is being directly adopted on the surface without firstly defining the exact location of edges. Thus, an approach has been carried out with the aim of developing the definition of edges to identify its exact location. The process of developing the definition of edges is conducted based on theoretical understanding and demonstration on the simulated surface topography data with the help of MATLAB software. Gradient-based edge detection operators are applied to see their performance on the defined surface feature boundary. Overall findings of this study showed that there are slight discrepancies of the edge position signified by the gradient-based operators to manually-defined edges. Despite of such outcome, it is then recommended to use either the Sobel operator or the Prewitt operator as a tool to signify the edge before proceeding with characterization process in future study due to their stability and clarity in producing the surface gradient function.

Impact of Multi-Feature Extraction on Image Retrieval and classification Using Machine Learning Technique

Vast information generated due to the web requires proper mechanisms and tools for efficient management and retrieval of images. This led to the development of an efficient image retrieval system. Image retrieval is the process of retrieving relevant images from a large database. Text-Based Image Retrieval (TBIR) and Content-Based Image Retrieval (CBIR) are the two efficient techniques for image retrieval. It has become an active area of research for image retrieval. Most of the systems are designed with three basic phases in CBIR, i.e. feature extraction, feature selection and similarity matching. Feature extraction can be done based on the colour, shape or texture of the image. The studies conducted so far involve CBIR systems which utilize individual features or only two or more features combined for feature extraction by applying different feature extraction and CBIR techniques. The proposed system illustrates a comparison between the performance of the existing CBIR algorithm with multi-feature extraction using discrete wavelet transform, edge histogram descriptor, sobel operator, moment invariant, histogram-oriented gradient, local binary pattern and classification using Support Vector Machine (SVM) based on the accuracy measures—precision and recall. The experimental results show that multi-feature extraction techniques give better results in some of the individual categories but retrieval results using machine learning is better with respect to average accuracy.

Development of Hybrid Automatic Segmentation Technique of a Single Leaf from Overlapping Leaves Image

The segmentation of a single leaf from an image with overlapping leaves is an important step towards the realization of effective precision agricultural systems. A popular approach used for this segmentation task is the hybridization of the Chan-Vese model and the Sobel operator CV-SO. This hybridized approach is popular because of its simplicity and effectiveness in segmenting a single leaf of interest from a complex background of overlapping leaves. However, the manual threshold and parameter tuning procedure of the CV-SO algorithm often degrades its detection performance. In this paper, we address this problem by introducing a dynamic iterative model to determine the optimal parameters for the CV-SO algorithm, which we dubbed the Dynamic CV-SO (DCV-SO) algorithm. This is a new hybrid automatic segmentation technique that attempts to improve the detection performance of the original hybrid CV-SO algorithm by reducing its mean error rate. The results obtained via simulation indicate that the proposed method yielded a 1.23% reduction in the mean error rate against the original CV-SO method.

Sobel Edge Detection Based on Weighted Nuclear Norm Minimization Image Denoising

As a classic and effective edge detection operator, the Sobel operator has been widely used in image segmentation and other image processing technologies. This operator has obvious advantages in the speed of extracting the edge of images, but it also has the disadvantage that the detection effect is not ideal when the image contains noise. In order to solve this problem, this paper proposes an optimized scheme for edge detection. In this scheme, the weighted nuclear norm minimization (WNNM) image denoising algorithm is combined with the Sobel edge detection algorithm, and the excellent denoising performance of the WNNM algorithm in a noise environment is utilized to improve the anti-noise performance of the Sobel operator. The experimental results show that the optimization algorithm can obtain better detection results when processing noisy images, and the advantages of the algorithm become more obvious with the increase of noise intensity.

Impact of noise to edge detection on images of different complexity

In this paper, an analysis of the edge detection over images of different complexity affected by Salt and Pepper, Gaussian and Speckle noise is performed. An analysis was performed for three noise levels, 0.01, 0.05 and 0.1. Over 100 images from the BSD database were used for analysis and each image has a GroundTruth with which an objective assessment of the detected edges was performed using PR and F measures. Five edge detectors Canny, LoG, Sobel, Prewitt and Roberts operator were used. The results are presented graphically. The obtained results show that noise significantly affects the detection of edges. When it comes to Salt and Pepper noise, Canny detector has achieved the best results for all levels of noise and image complexity. With the Speckle noise type for high and medium number of details in the image, Canny also gave the best results, while for low number of details in the image it is the Prewitt operator. When it comes to Gaussian noise, for all three categories of image complexity the best operator is Prewitt.

Segmentation and Features Extraction of Malaysian Herbs Leaves

In modern medical practice, herbs play a very important role as the source of bio-technology. Therefore, providing a database containing information of these herbs is significant to assist the medicinal practitioners and users. Image recognition, a process that includes image segmentation, features extraction and classification, is one of the methods to develop the database for herbs. This paper focuses on image segmentation and features extraction of 125 images with 1616×1080 pixels of Malaysian Herb Leaves including Sirih, Mexican Mint, Rerama, Belalai Gajah and Senduduk. The images were first segmented using Sobel operator to get the boundary points. Then, 7 geometrical features and 7 textural features were extracted for each image including area, perimeter, major axis length, minor axis length, roundness, smoothness and flatness. The experimental results were obtained using MATLAB R2018a. The results show that Sobel can successfully segment the images and calculates the features of the herb leaves.

Application of the tilt derivative transform to bathymetric data for structural lineament mapping

High-resolution bathymetry surveys provide an opportunity to analyse local geological structure where onshore areas afford limited exposure. Semi-automated lineament detection methods are necessary for areas of large coverage where a manual analysis would be subjective and time-consuming. However, semi-automated approaches are dependent on effective feature extraction methods to identify genuine lineaments. This study offers solutions to common problems that can impede processing methods where sharp steps in the seafloor (e.g. palaeocoastlines) are present. Directional gradient, Sobel and Laplacian filters are explored as well as the hillshade and tilt derivative transform for feature extraction prior to applying an object-based image analysis lineament detection approach. The filtered datasets generally perform poorly with a marked improvement when using the hillshade transform. However, we find the azimuth-invariant tilt derivative, which incorporates a convolved vertical derivative, to be the most successful, identifying lineaments in a range of orientations and across a sharp step in the seafloor.

Seeing the Vibration: Visual-Based Detection of Low Frequency Vibration Environment Pollution

Visual information can be easily obtained nowadays with more access to various photographic facilities. Appropriate logical analysis and processing of massive visual information could apply to flexible environment pollution detection. In this paper, we present a visual-based non-contact sensing method to detect the low frequency vibration pollution in daily life. The edge points of vibrating objects with more vibration information are extracted using bilateral filtering and Sobel operator. The Sobel operator weights the influence of the pixel position, which make the edge feature extraction effect better and the edge feature obtained more obvious. After that, the continuous frames of the vibration objects with edge information are transformed into several one-dimensional pixel level signals as the input of multi-scale network, which also improves the generalization ability. Experiments on daily objects under low frequency vibration are conducted. The detection accuracy is above 99%, which proves the reliability of the proposed method for the detection of low frequency vibration circumstance pollution.

RETRACTED ARTICLE: An improved convolutional neural network for abnormality detection and segmentation from human sperm images

Recent days, the infertility is affecting one of every ten couples. This makes the negative effect on the quality of a couple’s life, social causes, and psychological problems. Here the sperm morphology analysis helps to diagnose this problem. Here the machine learning approach is used for classification, detection and segmentation process. This also utilizes morphology approach for image representation. In this proposed method, the deep convolutional neural network is used for detecting the abnormality of human male infertility. Here the image morphological process is employed with the enhanced Otsu’s threshold method for segmenting the sperm image, which helps to detect the abnormal region using convolution layer. Here the database is collected from the human sperm image analysis dataset. Initially, the morphological process is applied to reduce the noise from the given set of input image then the segmentation process is performed by using E-Otsu’s threshold method. Two-dimensional Otsu’s thresholding technique reduces the computation complexity and it uses the median filter and for edge reduction approach sobel operator is used, which improves the performance of segmentation. Overall, the proposed research work optimizes three sections that are image representation by morphology approach, image segmentation by Enhanced-Otsu’s thresholding approach, and abnormality detection by Convolutional Neural Network. This method obtains the result of accuracy, detection rate, and computation time. By comparing with the existing method, the proposed method achieves the 98.99% of accuracy result and detects the abnormality effectively with the reduced computation time of 4 min and 15 s. This proposed work is done by using MATLAB with the adaptation of 2018a.

Efficient numerical modelling of time-domain light propagation in curved 3D absorbing and scattering media with finite differences.

An efficient approach is introduced for modelling light propagation in the time domain in 3D heterogeneous absorbing and scattering media (e.g. biological tissues) with curved boundaries. It relies on the finite difference method (FDM) in conjuction with the Crank-Nicolson method for accurately solving the optical diffusion equation (DE). The strength of the FDM lies in its simplicity and efficiency, since the equations are easy to set up, and accessing neighboring grid points only requires simple memory operations, leading to efficient code execution. Owing to its use of Cartesian grids, the FDM is generally thought cumbersome compared to the finite element method (FEM) for dealing with media with curved boundaries. However, to apply the FDM to such media, the blocking-off method can be resorted to. To account for the change of the refractive index at the boundary, Robin-type boundary conditions are considered. This requires the computation of surface normals. We resort here for the first time to the Sobel operator borrowed from image processing to perform this task. The Sobel operator is easy to implement, fast, and allows obtaining a smooth field of normal vectors along the boundary. The main contribution of this work is to arrive at a complete numerical FDM-based model of light propagation in the time domain in 3D absorbing and scattering media with curved geometries, taking into account realistic refractive index mismatch boundary conditions. The fluence rate obtained with this numerical model is shown to reproduce well that obtained with independent gold-standard Monte Carlo simulations.

Implementation of Sobel filter using CUDA

The Sobel filter is an edge detection filter that results in highlighting and focusing the edges. It is used for extracting the image gradient and contours in image processing applications. The data sets we are using in the paper are PGM and TIFF images with different rows and column pixel dimensions. The sobel filter operations used are implemented in C and CUDA programming language and the time taken for the operations in the serial and parallel platforms respectively are analyzed.

The Classification of Medicinal Plant Leaves Based on Multispectral and Texture Feature Using Machine Learning Approach

This study proposes the machine learning based classification of medical plant leaves. The total six varieties of medicinal plant leaves-based dataset are collected from the Department of Agriculture, The Islamia University of Bahawalpur, Pakistan. These plants are commonly named in English as (herbal) Tulsi, Peppermint, Bael, Lemon balm, Catnip, and Stevia and scientifically named in Latin as Ocimum sanctum, Mentha balsamea, Aegle marmelos, Melissa officinalis, Nepeta cataria, and Stevia rebaudiana, respectively. The multispectral and digital image dataset are collected via a computer vision laboratory setup. For the preprocessing step, we crop the region of the leaf and transform it into a gray level format. Secondly, we perform a seed intensity-based edge/line detection utilizing Sobel filter and draw five regions of observations. A total of 65 fused features dataset is extracted, being a combination of texture, run-length matrix, and multi-spectral features. For the feature optimization process, we employ a chi-square feature selection approach and select 14 optimized features. Finally, five machine learning classifiers named as a multi-layer perceptron, logit-boost, bagging, random forest, and simple logistic are deployed on an optimized medicinal plant leaves dataset, and it is observed that the multi-layer perceptron classifier shows a relatively promising accuracy of 99.01% as compared to the competition. The distinct classification accuracy by the multi-layer perceptron classifier on six medicinal plant leaves are 99.10% for Tulsi, 99.80% for Peppermint, 98.40% for Bael, 99.90% for Lemon balm, 98.40% for Catnip, and 99.20% for Stevia.

Grey Relevancy Degree and Improved Eight-Direction Sobel Operator Edge Detection

Edge detection is an important aspect to improve image edge quality in image processing. The purpose of edge detection is to identify the points in digital images with great brightness variation. However, the accuracy of traditional edge detection methods in edge extraction is low. For the actual image, the grey edge is sometimes not very clear, the image also contains noise. The detection result of the traditional Sobel operator is relatively accurate, but the detection result is rough and sensitive to noise. To solve the above problems, this paper proposes an improved eight-direction Sobel operator based on grey relevancy degree, which combines 5 × 5 Sobel operator with a grey relational degree and a new eight-direction grey relevancy method. The results show that this method can detect the useful information of edge more accurately and improve the anti-noise performance. However, the drawback is that the algorithm is not automatic.

DCNN-based Ship Classification using Enhanced Edge Information and Inception Module

The excellent feature extraction ability of deep convolutional neural networks (DCNNs) has been demonstrated in many image processing tasks, by which image classification can achieve high accuracy with only raw input images. However, the specific image features that influence the classification results are not readily determinable and what lies behind the predictions is unclear. This study proposes a method combining the Sobel and Canny operators and an Inception module for ship classification. The Sobel and Canny operators obtain enhanced edge features from the input images. A convolutional layer is replaced with the Inception module, which can automatically select the proper convolution kernel for ship objects in different image regions. The principle is that the high-level features abstracted by the DCNN, and the features obtained by multi-convolution concatenation of the Inception module must ultimately derive from the edge information of the preprocessing input images. This indicates that the classification results are based on the input edge features, which indirectly interpret the classification results to some extent. Experimental results show that the combination of the edge features and the Inception module improves DCNN ship classification performance. The original model with the raw dataset has an average accuracy of 88.72%, while when using enhanced edge features as input, it achieves the best performance of 90.54% among all models. The model that replaces the fifth convolutional layer with the Inception module has the best performance of 89.50%. It performs close to VGG-16 on the raw dataset and is significantly better than other deep neural networks. The results validate the functionality and feasibility of the idea posited.

Classification of Thorax X-ray Results on Corona Virus Infection Based on Internet of Things (IoT)

The spread of the corona virus in Indonesia is still growing until now. One important thing to note is about the handling of patients exposed to the coronavirus. This study aims to be able to classify thorax x-ray images against corona virus infection based on the Internet of Things (IoT). IoT is a concept of internet connectivity that is connected continuously between remote devices. This study uses a supervised learning method, namely directed learning in which the expected results of the user have been trained in advance and the information is stored in the system (database). The training data used is a Thorax x-ray image that is processed based on image processing with segmentation and edge detection based on the Sobel operator with real time connected to the internet. The results showed that the 100 times experiment obtained the accuracy value of the normal Thorax X-ray image of 94% and the Thorax covid-19 X-ray image of 96%. Overall system test showed 95% on target and 5% error.

Digital media zero watermark copyright protection algorithm based on embedded intelligent edge computing detection.

With the rapid development of computer technology and network communication technology, copyright protection caused by widely spread digital media has become the focus of attention in various fields. For digital media watermarking technology research emerge in endlessly, but the results are not ideal. In order to better realize the copyright identification and protection, based on the embedded intelligent edge computing detection technology, this paper studies the zero watermark copyright protection algorithm of digital media. Firstly, this paper designs an embedded intelligent edge detection module based on Sobel operator, including image line buffer module, convolution calculation module and threshold processing module. Then, based on the embedded intelligent edge detection module, the Arnold transform of image scrambling technology is used to preprocess the watermark, and finally a zero watermark copyright protection algorithm is constructed. At the same time, the robustness of the proposed algorithm is tested. The image is subjected to different proportion of clipping and scaling attacks, different types of noise, sharpening and blur attacks, and the detection rate and signal-to-noise ratio of each algorithm are calculated respectively. The performance of the watermark image processed by this algorithm is evaluated subjectively and objectively. Experimental data show that the detection rate of our algorithm is the highest, which is 0.89. In scaling attack, the performance of our algorithm is slightly lower than that of Fourier transform domain algorithm, but it is better than the other two algorithms. The Signal to Noise Ratio of the algorithm is 36.854% in P6 multiplicative noise attack, 39.638% in P8 sharpening edge attack and 41.285% in fuzzy attack. This shows that the algorithm is robust to conventional attacks. The subjective evaluation of 33% and 39% of the images is 5 and 4. The mean values of signal to noise ratio, peak signal to noise ratio, mean square error and mean absolute difference are 20.56, 25.13, 37.03 and 27.64, respectively. This shows that the watermark image processed by this algorithm has high quality. Therefore, the digital media zero watermark copyright protection algorithm based on embedded intelligent edge computing detection is more robust, and its watermark invisibility is also very superior, which is worth promoting.

Quantum image edge detection using improved Sobel mask based on NEQR

Recently, image edge detection using quantum image processing has focused due to having less circuit complexity and storage capacity. Edge extraction using the Sobel operator has the restriction in vertical and horizontal directions, which provides less edge information. In this paper, we introduce a quantum improved Sobel edge detection algorithm with non-maximum suppression and double threshold techniques for novel enhanced quantum representation method. We have analyzed the quantum circuit of realizing the edge detection algorithm, the number of edge pixels, simulation results, and circuit complexity. Thereafter, we have compared with the classical and some existing quantum edge detection algorithms. Our proposed algorithm can achieve a significant improvement in the case of edge information and circuit complexity.

An efficient enhanced automatic target recognition approach

This paper presents a new efficient enhanced automatic target recognition approach based on hybrid feature and intensity based methods. The proposed approach is based on comparison and matching between the acquired target image and the stored database images through edge detection, morphological operation, image transformation, image histogram matching and mutual information for both target and stored database images. Sobel operator is used to detect edges in both acquired target image and the stored database images. Hough transform is used as an image transformation which groups the extracted edge points in edged images to an appropriate set of lines which helps in features extraction and matching processes in both of target and stored database images. This gives an initial indication about realization and recognition between target image and its corresponding database image. Comparisons of image histogram and mutual information for both target and database images are finally used to emphasize the recognition of the target image and its verification with its corresponding database image. The proposed recognition approach passed through four stages which are: edge detection by Sobel edge detector, thinning as a morphological operation, Hough transformation, matching process using median deviation and finally histogram matching and measuring the mutual information between target and the available database images. The experimental results proved that the proposed hybrid target recognition approach gives a more accurate and successful recognition rate than other techniques which depend on using feature based methods or intensity based methods.

Validation Approach of the Tectonic Lineament Extraction Enhancement Using Sentinel 2A Images and Modified 3 × 3 Bidirectional Prewitt Filters. Case Study: Grombalia, Tunisia

In this study, filters are used to extract directional tectonic lineaments and, thus, to reveal the presence of faults or fractures in the satellite images. The extraction process consists first in eliminating other different types of lineaments (e.g. the lithological limit, ridgelines, hydrographic network, roads, etc.). Besides, a comparative and quantitative approach is applied to show that when the directional Prewitt filter has a zero setting of the opposite of the assumed direction, as is the case of the N-S direction, the East-West direction is also set to zero. This research work, based on a bidirectional filter (N-S; E-W), shows satisfactory results especially concerning the quantitative fluctuation of lineament directions by interval. This fluctuation is in perfect agreement with the lineaments provided by the digitizing of the different tectonic accidents directions extracted in a GIS environment from the geological map of Grombalia. In this research work, a quantitative approach was used to evaluate the result of the lineament extraction methodology based on one direction analysis and by an interval of directions. Indeed, the N45 lineament direction, well documented in the geological study of the region, was more clearly distinguished by applying the Prewitt filter than by using the Sobel filter. The result was validated by comparison with the results obtained by the digitization of the tectonic accidents mapped on the geological map.