Gabor Filters Research Articles

A robust and consistent stack generalized ensemble-learning framework for image segmentation

In the present study, we aim to propose an effective and robust ensemble-learning approach with stacked generalization for image segmentation. Initially, the input images are processed for feature extraction and edge detection using the Gabor filter and the Canny algorithms, respectively; our main goal is to determine the most feature descriptions. Subsequently, we applied the stacking generalization technique, which is generally built with two main learning levels. The first level is composed of two algorithms that give good results in the literature, namely: LightGBM (Light Gradient Boosting Machine) and SVM (support vector machine). The second level is the meta-model in which we use a predictor model that takes the base-level predictions to improve the accuracy of the final prediction. In the stacked generalization process, we use the Extreme Gradient Boosting (XGBoost); it takes as input the sub-models’ outputs to better classify each pixel of the image to give the final prediction. Today, several research works exist in the literature using different machine learning algorithms; in fact, instead of trying to find a single efficient and optimal learner, ensemble-based techniques take the advantage of each basic model; they integrate their outputs to obtain a more consistent and reliable learner. The result obtained from the models of individuals and our proposed approach is compared using a set of evaluation measures for image quality such as IoU, DSC, CC, SSIM, SAM, and UQI. The evaluation and a comparison of the results obtained showed more consistent predictions for the proposed model. Thus, we have made a comparison with some recent deep learning-based unsupervised segmentation methods. The evaluation and a comparison of the results obtained showed more coherent predictions for our stacked generalization in terms of precision, robustness, and consistency.

Fabric defect detection using ACS-based thresholding and GA-based optimal Gabor filter

Fabric defect detection is closely related to quality control. Therefore, it is an important topic in the modern textile field. In this article, a new fabric defect detection method using ACS-based thresholding and GA-based optimal Gabor filter is proposed. The multi-level gray-scale image thresholding is improved by using Adaptive cuckoo search (ACS) to calculate the best threshold. In order to suppress texture interference, the optimal Gabor filter calculated by genetic algorithm (GA) is used to achieve the optimal feature extraction of fabric defects. A specific image acquisition system is set up to obtain clear defective fabric images and corresponding defect-free images which are used to build the database. The performance of the method is evaluated by extensive experiments on various types of fabric defects. Experimental results show that the proposed method, effective and robust, is superior than the other six methods.

An Efficient Preprocessing Technique for Multimodality Breast Cancer Images

On average, one in every eight women is diagnosed with breast cancer during their lifetime, and accounts for 14% of cancers in women. Since early diagnosis could improve treatment outcomes and longer survival times for patients, it is absolutely necessary to develop techniques to classify lesions within breast cancer mammograms and ultrasound images. The main goal is to determine the class of tumor present within the image, which is pivotal in diagnosing breast cancer patients. In this paper, we propose an Sobel-Canny-Gabor(SCG) model, which is a hybrid model that implements three different edge detection filters; Sobel filter, Gabor filter, and Canny filter. This model is used to enhance the appearance of the mammogram and ultrasound images, which is then fed into a classification model. Through classification, there could be a potential improvement in the results of the overall classification. Post-classification, the model is then evaluated using the metric Peak Signal-to-Noise Ratio (PSNR), which measures the quality between the original image and the compressed image.On average, one in every eight women is diagnosed with breast cancer during their lifetime, and accounts for 14% of cancers in women. Since early diagnosis could improve treatment outcomes and longer survival times for patients, it is absolutely necessary to develop techniques to classify lesions within breast cancer mammograms and ultrasound images. The main goal is to determine the class of tumor present within the image, which is pivotal in diagnosing breast cancer patients. In this paper, we propose an Sobel-Canny-Gabor(SCG) model, which is a hybrid model that implements three different edge detection filters; Sobel filter, Gabor filter, and Canny filter. This model is used to enhance the appearance of the mammogram and ultrasound images, which is then fed into a classification model. Through classification, there could be a potential improvement in the results of the overall classification. Post-classification, the model is then evaluated using the metric Peak Signal-to-Noise Ratio (PSNR), which measures the quality between the original image and the compressed image.

Field path optimization to reduce headland and turning maneuvers at regional scales: automated detection of cultivation direction in the state of Brandenburg, Germany

Path planning for optimized field-work pattern is an important task within precision farming. The decision on a particular direction and path to cultivate and manage the field is complex and can significantly affect working time, energy consumption, soil compaction and yield. This study proposed a new method for automated detection of the current cultivation direction of several thousands of agricultural fields and compared the current cultivation direction with an optimized cultivation direction generated from a path planning algorithm. Airborne imagery from 2019 was analyzed using a modified Gabor filter. The identification takes place on a sub-plot level and can therefore detect small-scale differences in cultivation direction within fields. The method for identification of current cultivation direction had a high success rate of 87.5%. Fields with a high potential to save turning maneuvers and to reduce the area of headland were identified. From 3410 fields, a total of 58162 turning maneuvers and 507 ha headland were saved. This corresponds to 14.1% of all turning maneuvers and 7.6% of the total headland area for all analyzed fields in Brandenburg. A high optimization potential was demonstrated for field paths when efficient processing directions are taken into account. The method can be extended to the analysis of satellite imagery and thus offers the possibility of identifying current cultivation directions with a high spatial and temporal resolution. In future, this knowledge can be embedded within decision support systems for real-time optimization of field machinery path planning to support sustainable cropping practices.

Gabor frames from contact geometry in models of the primary visual cortex

We analyze the interplay between contact geometry and Gabor filters signal analysis in geometric models of the primary visual cortex. We show in particular that a specific framed lattice and an associated Gabor system is determined by the Legendrian circle bundle structure of the $3$-manifold of contact elements on a surface (which models the V1-cortex), together with the presence of an almost-complex structure on the tangent bundle of the surface (which models the retinal surface). We identify a scaling of the lattice, also dictated by the manifold geometry, that ensures the frame condition is satisfied. We then consider a $5$-dimensional model where receptor profiles also involve a dependence on frequency and scale variables, in addition to the dependence on position and orientation. In this case we show that a proposed profile window function does not give rise to frames (even in a distributional sense), while a natural modification of the same generates Gabor frames with respect to the appropriate lattice determined by the contact geometry.

EMOTİON RECOGNİTİON USİNG EXPLİCİT FEATURE EXTRACTİON METHOD

Emotions are conveyed every single second, and understanding what they typically represent is crucial in practically every industry. People's ability to accurately recognise emotions varies greatly, making technology utilized for the purpose a relatively new subject of research. We attempted to work on extracting significant features from the image datasets and feeding those characteristics to a CNN in this study to aid the emotion detection for the CKplus and JAFFE dataset. The two separate extraction techniques employed are the Gabor filter and SIFT. The extraction techniques assist in identifying the main attributes that contain a profusion of visual data.The experimental investigation demonstrates that the gabor filter paired with CNN outperforms the state-of-the-art performance using the CKplus dataset, with an overall accuracy of 98.12.

Spectrum sensing in cognitive radio networks using an ensemble of machine learning frameworks and effective feature extraction

From the signal received on a particular frequency band, spectrum sensing (SS) is used in cognitive radio (CR) to assess whether the primary user (PU) is using the spectrum and, consequently, whether the secondary user (SU) can utilize the spectrum. The main issue with SS is determining the presence of the primary signal in a low signal-to-noise ratio (SNR). Compared to conventional technologies, machine learning techniques are more effective and accurate at identifying the qualities of input data. This paper proposes a machine learning (ML) based SS model for CR with effective feature extraction and reduction techniques. The proposed work comprises five phases: noise removal, wavelet transform, feature extraction, dimensionality reduction, and classification. Firstly, noise filtering is done on the received signal to remove the noise present in the input signal using the filters such as moving median filter (MMF), Gaussian filter (GF), and Gabor filter (GBF). After that, the filtered signal is transformed into a wavelet domain using Discrete Wavelet Transform (DWT) algorithm. Then the statistical features such as average absolute value, wavelet energy, variance, standard deviation, and peak value features are extracted from the DWT. Next, the dimensionality reduction (DR) is performed using Linear Discriminant Analysis (LDA). Finally, the classification is performed using the ensemble ML classifiers such as Support Vector Machine (SVM), Naive Bayes (NB), and K-Nearest Neighbour (KNN), which classify whether the PU signal is active or not. Simulations are carried out to analyze the efficiency of the presented models for SS. The results proved that SVM obtains the best performance for SS with higher accuracy and lower SNR.

A New Region-Based Minimal Path Selection Algorithm for Crack Detection and Ground Truth Labeling Exploiting Gabor Filters

Cracks are fractures or breaks that occur in materials such as concrete, metals, rocks, and other solids. Various methods are used to detect and monitor cracks; among many of them, image-based methodologies allow fast identification of the distress and easy quantification of the percentage of cracks in the scene. Two main categories can be identified: classical and deep learning approaches. In the last decade, the tendency has moved towards the use of the latter. Even though they have proven their outstanding predicting performance, they suffer some drawbacks: a “black-box” nature leaves the user blind and without the possibility of modifying any parameters, a huge amount of labeled data is generally needed, a process that requires expert judgment is always required, and, finally, they tend to be time-consuming. Accordingly, the present study details the methodology for a new algorithm for crack segmentation based on the theory of minimal path selection combined with a region-based approach obtained through the segmentation of texture features extracted using Gabor filters. A pre-processing step is described, enabling the equalization of brightness and shadows, which results in better detection of local minima. These local minimal are constrained by a minimum distance between adjacent points, enabling a better coverage of the cracks. Afterward, a region-based segmentation technique is introduced to determine two areas that are used to determine threshold values used for rejection. This step is critical to generalize the algorithm to images presenting close-up scenes or wide cracks. Finally, a geometrical thresholding step is presented, allowing the exclusion of rounded areas and small isolated cracks. The results showed a very competitive F1-score (0.839), close to state-of-the-art values achieved with deep learning techniques. The main advantage of this approach is the transparency of the workflow, contrary to what happens with deep learning frameworks. In the proposed approach, no prior information is required; however, the statistical parameters may have to be adjusted to the particular case and requirements of the situation. The proposed algorithm results in a useful tool for researchers and practitioners needing to validate their results against some reference or needing labeled data for their models. Moreover, the current study could establish the grounds to standardize the procedure for crack segmentation with a lower human bias and faster results. The direct application of the methodology to images obtained with any low-cost sensor makes the proposed algorithm an operational support tool for authorities needing crack detection systems in order to monitor and evaluate the current state of the infrastructures, such as roads, tunnels, or bridges.

Rotational Invariance Using Gabor Convolution Neural Network and Color Space for Image Processing

Convolutional neural networks (CNNs) are deep learning methods that are utilized in image processing such as image classification and recognition. It has achieved excellent results in various sectors; however, it still lacks rotation invariant and spatial information. To establish whether two images are rotational versions of one other, one can rotate them exhaustively to see if they compare favorably at some angle. Due to the failure of current algorithms to rotate images and provide spatial information, the study proposes to transform color spaces and use the Gabor filter to address the issue. To gather spatial information, the HSV and CieLab color spaces are used, and Gabor is used to orient images at various orientation. The experiments show that HSV and CieLab color spaces and Gabor convolutional neural network (GCNN) improves image retrieval with an accuracy of 98.72% and 98.67% on the CIFAR-10 dataset.

Klasifikasi Jenis Jambu Air Berdasarkan Citra Daun Menggunakan Metode Filter Gabor Dan K-NN (K-Nearest Neighbors)

Water guava is a fruit that has many benefits. There are many types of water guava such as Leang, Citra, and Kaget. These types of water guava have some similarities in leaf shape, making it difficult for ordinary people to recognize the type of water guava plant. This study aims to make it easier for ordinary people to recognize the types of guava based on the texture of the leaves. The method used is the Gabor Filter and K-NN (K-Nearest Neighbor). Before extracting the leaf image data, a total of 150 leaf image datasets from each type of guava (Leang, Citra, Kaget) were converted to grayscale first and then extracted using a Gabor filter. Then classification is carried out with K-NN using the Gabor filter method and K-NN. The accuracy results obtained by combining the two methods are 93.34%.

Copy-Move Forgery Detection Using Texture Features of Hidden Forged Regions

The recent revolution in technology has not only eased our daily activities at work and home but also introduced new threats. In their daily activities, people exchange a lot of files such as text files, images, videos, etc. that can be used for a variety of purposes. One of the most common types of files is images. These kinds of files can be used to socialize people or spread knowledge among communities. Some of the exchanged images are fake or forged which can lead to the spread of misinformation, which is dangerous. This paper tries to suggest a method for image forgery detection that is copy-move-based. This means a part of the image is used to hide or change other parts in the same image. The suggested method divides an image into several blocks. The feature vectors of the blocks are extracted using a modified Gabor filter. The extracted features are, then, reduced using the principal component analysis technique. The next step is to match the blocks and extract similar ones (duplicated blocks). The findings show that the suggested method is efficient compared to other methods in the literature in terms of detection rate and false positive detection. Also, the proposed method detected forged regions of images when having a 60% of compression rate.

Smart fabric inspection using Mimosa pudica plant

Fabric quality governing and defect detection are playing a crucial role in the textile industry with the development of high customer demand in the fashion market. This work presents fabric defect detection using the sensitive plant segmentation algorithm (SPSA) which, is developed with the sensitive behaviour of the plant biologically named “Mimosa pudica”i. This method consists of two stages. The first stage enhances the contrast of the defective fabric image and the second stage segments the fabric defects with aid of SPSA. The proposed work SPSA is developed for defective pixels identification in both uniform and non-uniform patterns of fabrics. In this work, SPSA has been done by checking with devised condition, correlation and error probability. Every pixel will be checked with the developed algorithm, to get marked either defective or non-defective pixels. The proposed SPSA has been tested on the different types of fabric defect databases and shows a prodigious performance over existing methods like the Differential evolution based optimal Gabor filter model (DEOGF), Gabor filter bank (GFB), Adaptive sparse representation-based detection model (ASR) and Fourier and wavelet shrinkage (FWR).

Age-dependent changes in epidermal architecture explored using an automated image analysis algorithm on in vivo reflectance confocal microscopy images.

Reflectance confocal microscopy (RCM) allows for real-time in vivo visualization of the epidermis at the cellular level noninvasively. Parameters relating to tissue architecture can be extracted from RCM images, however, analysis of such images requires manual identification of cells to derive these parameters, which can be time-consuming and subject to human error, highlighting the need for an automated cell identification method. First, the region-of-interest (ROI) containing cells needs to be identified, followed by the identification of individual cells within the ROI. To perform this task, we use successive applications of Sato and Gabor filters. The final step is post-processing improvement of cell detection and removal of size outliers. The proposed algorithm is evaluated on manually annotated real data. It is then applied to 5345 images to study the evolution of epidermal architecture in children and adults. The images were acquired on the volar forearm of healthy children (3 months to 10 years) and women (25-80 years), and on the volar forearm and cheek of women (40-80 years). Following the identification of cell locations, parameters such as cell area, cell perimeter, and cell density are calculated, as well as the probability distribution of the number of nearest neighbors per cell. The thicknesses of the Stratum Corneum and supra-papillary epidermis are also calculated using a hybrid deep-learning method. Epidermal keratinocytes are significantly larger (area and perimeter) in the granular layer than in the spinous layer and they get progressively larger with a child's age. Skin continues to mature dynamically during adulthood, as keratinocyte size continues to increase with age on both the cheeks and volar forearm, but the topology and cell aspect ratio remain unchanged across different epidermal layers, body sites, and age. Stratum Corneum and supra-papillary epidermis thicknesses increase with age, at a faster rate in children than in adults. The proposed methodology can be applied to large datasets to automate image analysis and the calculation of parameters relevant to skin physiology. These data validate the dynamic nature of skin maturation during childhood and skin aging in adulthood.

Predicting Level Measurements by Supervised Learning Based on Gabor and Smote Filters: An Industrial Non-Interacting Tanks Scenario

Fluid - Level measurement is required in recognizing the state variable of a Level processing plant for monitoring the level deviations in most the industrial plants. This issue has mostly been addressed by conventional methods using level sensors and level transmitters. However, the cost associated with these mechanisms can be reduced. It has been identified the chances with the applications machine learning approach in this scenario. Such a solution first of its kind is proposed and tried in this article. Gabor feature selection is established with SMOTE (Synthetic Minority Oversampling Technique) to overcome the imbalanced data set to improve the accuracy performance using Weka software. This paper aims to construct a level predictor based on machine learning algorithms and to demonstrate the training and testing performance validated with more accuracy. At this juncture, we consider the fluid level images for analyzing and predicting the level measurement by building supervised models with selected classifiers like IBk, JRip, J48, and Random Forest machine learning algorithms. The java-implemented tool Weka is subjected to get the maximum accuracy of 76.4286% and weighted average ROC (Receiver Operator Characteristics) values of 0.903. Hence such smart measurements using Machine learning algorithms, part of Artificial intelligence provide us a vital role in measuring the level parameter in less time using image processing filters.

Application of SIFT Algorithm based on the Gabor Features in Multi-source Information Image Monitoring

Under the background of today’s intelligence, the application of intelligent images is becoming popularized, and the detection of multi-source information images is a valuable research topic. Current image detection methods are insensitive and inaccurate. Therefore, the research combines the SIFT (scale-invariant feature transform) algorithm with the Gabor features and CNN (convolutional neural network (Ed note. Acronyms only need to be defined once.)) to form an improved new algorithm. The algorithm divides the features of the image into several different categories. In each category, the features will be fully identified and extracted, and different levels of feature matching will be performed. The properties of the SIFT algorithm are used to form an operational stacking pyramid and combine the Gabor features and CNN with it. The Gabor filter is formed into a filter bank to obtain parameters, including frequency, scale, and direction in various dimensions. The results are fused to obtain a fusion Gabor descriptor. The high sensitivity of CNN to images, particularly colors, is applied to the algorithm to make monitoring the algorithm more accurate. The experimental results show that the average precision rate, average recall rate, and average precision rate of the improved algorithm are 92.35%, 74.79%, and 82.55%, respectively, which are significantly higher than the other two algorithms used for comparison. The improved algorithm shows better performance and has remarkable advantages that can be applied to the image monitoring of image information.

Cervical precancerous lesion classification using quantum invasive weed optimization with deep learning on biomedical pap smear images

AbstractBiomedical imaging devices, in general, have been made and used a lot lately to examine the insides of the body during diagnostic and analytic procedures. Biomedical imaging gives accurate information about metabolites, which can be used to find and classify diseases because it is not invasive. For the study of cervical cancer (CC), the pap smear is a crucial type of biological imaging. CC is a crucial reason to enhance the rate of women's mortalities. Proper screening of pap smear images is critical for assisting in the early detection and analysis of CC. Computer‐aided systems for cancerous cell recognition need well established artificial intelligence (AI) methods. In this study, we introduce an automated Cervical Precancerous Lesion Classification using Quantum Invasive Weed Optimization with Deep Learning (CPLC‐QIWODL) on biomedical pap smear images. The presented CPLC‐QIWODL technique examines the pap smear images for cervical cancer classification. To do so, the presented CPLC‐QIWODL technique pre‐processes the biomedical images using a Gabor filtering (GF) approach. Moreover, the CPLC‐QIWODL technique uses a deep convolutional neural network‐based SqueezeNet system for feature extraction. Furthermore, the hyperparameter tuning of the SqueezeNet methodology takes place using the QIWO technique, showing the novelty of the work. Finally, to classify CC, the deep variational autoencoder (DVAE) model is applied. The experimental result analysis of the CPLC‐QIWODL technique is tested using a benchmark medical image database. Extensive comparative results demonstrated the enhanced outcomes of the CPLC‐QIWODL technique over other existing algorithms, with a maximum accuracy of 99.07%.

A fusion approach to infrared and visible images with Gabor filter and sigmoid function

The fusion of infrared (IR) and visible images should not only increase the brightness of the infrared targets, but also preserve more details in visible images. A fusion approach to infrared and visible images with Gabor filter and sigmoid function is proposed in this paper. In order to make targets prominent in the fused image, the IR image is normalized by the sigmoid function to get a mapping matrix W, and then the visible image is enhanced by the matrix W to obtain an enhanced visible image, so that the contrast between the target and the background in the visible image is enhanced. After the decomposition of the IR image, the visible image and the enhanced visible image with Gabor filter, the detail layers of the visible image and the enhanced visible image are fused using the “max-absolute” rule, and the base layers are fused using the rule of weighted summation, which greatly increases the amount of information in the visible image. The weighted summation is calculated for the basic layers of the infrared and fused visible image, and the absolute maximum value is calculated for their detail layers. Finally, the fused image is generated by a linear combination of the final detail layer and the final base layer. Nine existing algorithms with better performance and the algorithm proposed in this paper are tested on public datasets, and use six evaluation metrics such as average gradient (AG), cross entropy (CE), edge gradient (EI), information entropy (IE), peak signal-to-noise ratio (PSNR) and spatial frequency (SF) to evaluate the quality of the fused image. Experiments show that the new algorithm has achieved better visual effects, and most of the objective evaluation metrics are also better than other algorithms. In particular, the low-light image is enhanced by the sigmoid function and then fused, so that the fused image is clearer and the target is more prominent. The gradient information is retained in the fused image as much as possible. All these prove the advantage and effectiveness of the new algorithm.

Gender Recognition of Human from Face Images Using Multi-Class Support Vector Machine (SVM) Classifiers

In the realm of robotics and interactive systems, gender recognition is a crucial problem. Considering the several uses it has in security, web search, human-computer interactions, etc., gender recognition from facial photos has garnered a lot of attention. The need to use and enhance gender recognition techniques is felt more strongly today due to a significant development in the design of facial recognition systems. Relatively speaking to other approaches, the progress gained in this area thus far is not exceptional. Thus, a novel method has been adopted in this study to improve accuracy in comparison to earlier research. To create the best rate of accuracy and efficiency in the suggested method of this research, we choose a minimal set of characteristics. Testing on the FERET and UTK-Face datasets reveals that our suggested algorithm has a lower degree of inaccuracy. In this article, the input image of the person's face is pre-processed to extract the right features from the face once the person's face has been recognized. Gender separation is achieved using Multi-class Support Vector Machine (SVM) Classifiers after features from normalized images have been extracted using Histogram Oriented Gradient (HOG), Gabor Filters, and Speeded Up Robust Features (SURF), as well as their combination to select the most appropriate feature from them as input for gender classification. As a feature reduction feature, the Principal Component Analysis (PCA) algorithm is also employed. Using the proposed approach, 98.75% gender recognition precision has been accomplished on the FERET database and a runtime performance of 0.4 Sec. on the UTK-Face database, 97.43% gender recognition accuracy has been accomplished and a runtime performance of 0.5 Sec.

Optimized Classification of Intelligent Reflecting Surface (IRS)-Enabled GEO Satellite Signals.

The intelligent reflecting surface (IRS) is a cutting-edge technology for cost-effectively achieving future spectrum- and energy-efficient wireless communication. In particular, an IRS comprises many low-cost passive devices that can independently reflect the incident signal with a configurable phase shift to produce three-dimensional (3D) passive beamforming without transmitting Radio-Frequency (RF) chains. Thus, the IRS can be utilized to greatly improve wireless channel conditions and increase the dependability of communication systems. This article proposes a scheme for an IRS-equipped GEO satellite signal with proper channel modeling and system characterization. Gabor filter networks (GFNs) are jointly proposed for the extraction of distinct features and the classification of these features. Hybrid optimal functions are used to solve the estimated classification problem, and a simulation setup was designed along with proper channel modeling. The experimental results show that the proposed IRS-based methodology provides higher classification accuracy than the benchmark without the IRS methodology.

An effective fingerprint orientation field estimation method using differential values of grayscale intensity.

Fingerprint orientation field (OF) estimation is important for basic fingerprint image processing and impacts the accuracy of fingerprint image enhancements, such as Gabor filters. In this article, we introduce an OF estimation algorithm based on differential values of grayscale intensity and examine the accuracy and reliability of the proposed algorithm by applying it to fingerprint images processed using Gaussian blurring and the Gaussian white noise process. The experimental results indicate that the OF estimation reliability of the proposed algorithm is higher than the gradient-based method and the power spectral density (PSD) based method in low quality fingerprints. The proposed algorithm is especially useful in noisy fingerprint images, where the OF estimation reliability of the algorithm is 6.46% and 32.93% higher than the gradient-based method and the PSD-based method, respectively.