Ternary Pattern Research Articles

Copy-Move Forgery Verification in Images Using Local Feature Extractors and Optimized Classifiers

Passive image forgery detection methods that identify forgeries without prior knowledge have become a key research focus. In copy-move forgery, the assailant intends to hide a portion of an image by pasting other portions of the same image. The detection of such manipulations in images has great demand in legal evidence, forensic investigation, and many other fields. The paper aims to present copy-move forgery detection algorithms with the help of advanced feature descriptors, such as local ternary pattern, local phase quantization, local Gabor binary pattern histogram sequence, Weber local descriptor, and local monotonic pattern, and classifiers such as optimized support vector machine and optimized NBC. The proposed algorithms can classify an image efficiently as either copy-move forged or authenticated, even if the test image is subjected to attacks such as JPEG compression, scaling, rotation, and brightness variation. CoMoFoD, CASIA, and MICC datasets and a combination of CoMoFoD and CASIA datasets images are used to quantify the performance of the proposed algorithms. The proposed algorithms are more efficient than state-of-the-art algorithms even though the suspected image is post-processed.

Reflectance material classification using optimized deep learning and change detection of LANDSAT surface reflectance images

The time-series changes that are affected by a specific alteration cannot be accurately found using change detection techniques. However, there are not many techniques available for identifying seasonal changes. This study develops a new deep model for identifying reflectance materials depending on optimization. At time instances “t” and “t-1” the images are accumulated in this instance. The Type 2 Fuzzy and Cuckoo Search-Based Filter (T2FCS) is used for the pre-processing of two images. Then, Fuzzy Local Information C-Means (FLICM) is used to segment both images. On the image accumulated at time “t” the feature extraction is performed to extract statistical features like mean, variance, kurtosis, energy, skewness, and standard deviation as well as imperative features like Local Optimal Oriented Pattern (LOOP), Local Ternary Pattern (LTP), Local Gabor XOR Pattern (LGXP), and others. The reflectance material is then categorized using a Deep Neuro-Fuzzy Network (DNFN) with segments created from images collected at time instance “t-1” and features collected from “t”. The proposed Jaya Student psychology-based optimization (JSPO), which was created by fusing the Jaya algorithm and the Student psychology-based optimization method(SPBO), is used here to train the DNFN. The segmented output that was collected at times “t” and “t-1” is then used to do the change detection. With the highest accuracy (94.1%) and maximum sensitivity (93.8%), the proposed JSPO-based DNFN demonstrated improved performance. For applications involving optical imagery, the Landsat program is crucial.

A novel fault classification feature extraction method for rolling bearing based on multi-sensor fusion technology and EB-1D-TP encoding algorithm

To improve the accuracy of bearing fault diagnosis in a multisensor monitoring environment, it is necessary to obtain more accurate and effective fault classification features for bearings. Accordingly, a bearing fault classification feature extraction method based on multisensor fusion technology and an enhanced binary one-dimensional ternary pattern (EB-1D-TP) algorithm were proposed in this study. First, an optimal equalization weighting algorithm was established to realize high-precision fusion of bearing signals by introducing an optimal equalization factor and determining the theoretical optimal equalization factor value. Second, an enhanced binary encoding method similar to balanced ternary encoding was developed, which increases the difference between the different fault features of the bearing. Finally, the new sequence obtained after encoding was used as the input to a support vector machine to classify and diagnose the faults of the rolling bearing. The experimental results show that the algorithm can significantly improve the accuracy and speed of rolling-bearing fault classification. Combining fusion-encoding features with other intelligent classification methods, the classification results were improved.

A novel higher order appearance texture analysis to diagnose lung cancer based on a modified local ternary pattern

Background and ObjectiveLung cancer is an important cause of death and morbidity around the world. Two of the primary computed tomography (CT) imaging markers that can be used to differentiate malignant and benign lung nodules are the inhomogeneity of the nodules’ texture and nodular morphology. The objective of this paper is to present a new model that can capture the inhomogeneity of the detected lung nodules as well as their morphology. MethodsWe modified the local ternary pattern to use three different levels (instead of two) and a new pattern identification algorithm to capture the nodule’s inhomogeneity and morphology in a more accurate and flexible way. This modification aims to address the wide Hounsfield unit value range of the detected nodules which decreases the ability of the traditional local binary/ternary pattern to accurately classify nodules’ inhomogeneity. The cut-off values defining these three levels of the novel technique are estimated empirically from the training data. Subsequently, the extracted imaging markers are fed to a hyper-tuned stacked generalization-based classification architecture to classify the nodules as malignant or benign. The proposed system was evaluated on in vivo datasets of 679 CT scans (364 malignant nodules and 315 benign nodules) from the benchmark Lung Image Database Consortium and Image Database Resource Initiative (LIDC-IDRI) and an external dataset of 100 CT scans (50 malignant and 50 benign). The performance of the classifier was quantitatively assessed using a Leave-one-out cross-validation approach and externally validated using the unseen external dataset based on sensitivity, specificity, and accuracy. ResultsThe overall accuracy of the system is 96.17% with 97.14% sensitivity and 95.33% specificity. The area under the receiver-operating characteristic curve was 0.98, which highlights the robustness of the system. Using the unseen external dataset for validating the system led to consistent results showing the generalization abilities of the proposed approach. Moreover, applying the original local binary/ternary pattern or using other classification structures achieved inferior performance when compared against the proposed approach. ConclusionsThese experimental results demonstrate the feasibility of the proposed model as a novel tool to assist physicians and radiologists for lung nodules’ early assessment based on the new comprehensive imaging markers.

Intracranial hemorrhage detection in 3D computed tomography images using a bi-directional long short-term memory network-based modified genetic algorithm.

Intracranial hemorrhage detection in 3D Computed Tomography (CT) brain images has gained more attention in the research community. The major issue to deal with the 3D CT brain images is scarce and hard to obtain the labelled data with better recognition results. To overcome the aforementioned problem, a new model has been implemented in this research manuscript. After acquiring the images from the Radiological Society of North America (RSNA) 2019 database, the region of interest (RoI) was segmented by employing Otsu's thresholding method. Then, feature extraction was performed utilizing Tamura features: directionality, contrast, coarseness, and Gradient Local Ternary Pattern (GLTP) descriptors to extract vectors from the segmented RoI regions. The extracted vectors were dimensionally reduced by proposing a modified genetic algorithm, where the infinite feature selection technique was incorporated with the conventional genetic algorithm to further reduce the redundancy within the regularized vectors. The selected optimal vectors were finally fed to the Bi-directional Long Short Term Memory (Bi-LSTM) network to classify intracranial hemorrhage sub-types, such as subdural, intraparenchymal, subarachnoid, epidural, and intraventricular. The experimental investigation demonstrated that the Bi-LSTM based modified genetic algorithm obtained 99.40% sensitivity, 99.80% accuracy, and 99.48% specificity, which are higher compared to the existing machine learning models: Naïve Bayes, Random Forest, Support Vector Machine (SVM), Recurrent Neural Network (RNN), and Long Short-Term Memory (LSTM) network.

3D Road Lane Classification with Improved Texture Patterns and Optimized Deep Classifier.

The understanding of roads and lanes incorporates identifying the level of the road, the position and count of lanes, and ending, splitting, and merging roads and lanes in highway, rural, and urban scenarios. Even though a large amount of progress has been made recently, this kind of understanding is ahead of the accomplishments of the present perceptual methods. Nowadays, 3D lane detection has become the trending research in autonomous vehicles, which shows an exact estimation of the 3D position of the drivable lanes. This work mainly aims at proposing a new technique with Phase I (road or non-road classification) and Phase II (lane or non-lane classification) with 3D images. Phase I: Initially, the features, such as the proposed local texton XOR pattern (LTXOR), local Gabor binary pattern histogram sequence (LGBPHS), and median ternary pattern (MTP), are derived. These features are subjected to the bidirectional gated recurrent unit (BI-GRU) that detects whether the object is road or non-road. Phase II: Similar features in Phase I are further classified using the optimized BI-GRU, where the weights are chosen optimally via self-improved honey badger optimization (SI-HBO). As a result, the system can be identified, and whether it is lane-related or not. Particularly, the proposed BI-GRU + SI-HBO obtained a higher precision of 0.946 for db 1. Furthermore, the best-case accuracy for the BI-GRU + SI-HBO was 0.928, which was better compared with honey badger optimization. Finally, the development of SI-HBO was proven to be better than the others.

Histogram Refined Local Ternary Pattern-based Bilateral LPP for Vision Sensor-based Robot Navigation Guidance under Challenging Environments

The present paper proposes a new variant of bilateral locality preserving projection (BLPP), a nonlinear manifold learning technique, for vision sensor-based robot navigation guidance in challenging environments, suffering from both variations in ambient illumination levels and vision sensor noises and uncertainties in image acquisition. In its basic form, the adjacency graph in BLPP is formulated both in feature subspace and spatial subspace, simultaneously. The present work proposes to compute the feature weights using local ternary pattern (LTP), a lesser noise-sensitive feature descriptor, which is also robust against illumination variations. In addition, histogram refinement has been performed on the feature descriptor to encode the local neighborhood structure of the data, proposing a further augmented version of BLPP named histogram refined local ternary pattern-based BLPP (HRLTP-BLPP). Extensive real-life experiments with challenging photometric conditions and sensor noises have been performed to show the usefulness of HRLTP-BLPP over other competing, state-of-the-art approaches, utilized for robot navigation guidance.

Image Retrieval System Using Kirsch based Local Ternary Pattern

Image Retrieval System Using Kirsch based Local Ternary Pattern

Opponent colour median-based scale invariant local ternary pattern: a new feature descriptor for moving object detection

ABSTRACT This article presents a new colour-texture feature descriptor called opponent colour median-based scale invariant local ternary pattern (OCMD-SILTP) to deal with dynamic backgrounds, similar texture regions, moving cast shadow, and illumination changes. Here, the incoming image frames are first divided into several sub regions and then the colour-texture features are derived by applying OCMD-SILTP on the individual colour planes (R, G, and B) as well as opponent colour planes (R-G, R-B, and G-B). For opponent planes, the median of neighbourhood pixels of one color plane is used for thresholding the neighbourhood pixels of opponent color plane. Because of this, the proposed OCMD-SILTP colour-texture descriptors are more immune to scaling and video noises. The background changes are dynamically updated through OCMD-SILTP texture patterns using histogram intersection method. Experiments on publically available dataset prove the significance of proposed approach under complex environments.

A Novel ST-ViBe Algorithm for Satellite Fog Detection at Dawn and Dusk

Satellite remote sensing provides a potential technology for detecting fog at dawn and dusk on a large scale. However, the spectral characteristics of fog at dawn and dusk are similar to those of the ground surface, which makes satellite-based fog detection difficult. With the aid of time-series datasets from the Himawari-8 (H8)/AHI, this study proposed a novel algorithm of the self-adaptive threshold of visual background extractor (ST-ViBe) model for satellite fog detection at dawn and dusk. Methodologically, the background model was first built using the difference between MIR and TIR (BTD) and the local binary similarity patterns (LBSP) operator. Second, BTD and scale invariant local ternary pattern (SILTP) texture features were coupled to form scene factors, and the detection threshold of each pixel was determined adaptively to eliminate the influence of the solar zenith angles. The background model was updated rapidly by accelerating the updating rate and increasing the updating quantity. Finally, the residual clouds were removed with the traditional cloud removal method to achieve accurate detection of fog at dawn and dusk over a large area. The validation results demonstrated that the ST-ViBe algorithm could detect fog at dawn and dusk precisely, and on a large scale. The probability of detection, false alarm ratio, and critical success index were 72.5%, 18.5%, 62.4% at dawn (8:00) and 70.6%, 33.6%, 52.3% at dusk (17:00), respectively. Meanwhile, the algorithm mitigated the limitations of the traditional algorithms, such as illumination mutation, missing detection, and residual shadow. The results of this study could guide satellite fog detection at dawn and dusk and improve the detection of similar targets.

Feature Extraction Method using HoG with LTP for Content-Based Medical Image Retrieval

An accurate diagnosis is significant for the treatment of any disease in its early stage. Content-Based Medical Image Retrieval (CBMIR) is used to find similar medical images in a huge database to help radiologists in diagnosis. The main difficulty in CBMIR is semantic gaps between the lower-level visual details, captured by computer-aided tools and higher-level semantic details captured by humans. Many existing methods such as Manhattan Distance, Triplet Deep Hashing, and Transfer Learning techniques for CBMIR were developed but showed lower efficiency and the computational cost was high. To solve such issues, a new feature extraction approach is proposed using Histogram of Gradient (HoG) with Local Ternary Pattern (LTP) to automatically retrieve medical images from the Contrast-Enhanced Magnetic Resonance Imaging (CE-MRI) database. Adam optimization algorithm is utilized to select features and the Euclidean measure calculates the similarity for query images. From the experimental analysis, it is clearly showing that the proposed HoG-LTP method achieves higher accuracy of 98.8%, a sensitivity of 98.5%, and a specificity of 99.416%, which is better when compared to the existing Random Forest (RF) method which displayed an accuracy, sensitivity, and specificity of 81.1%, 81.7% and 90.5% respectively.

Classification of intracranial hemorrhage CT images based on texture analysis using ensemble-based machine learning algorithms: A comparative study

Blood vessels in the brain tissue can leak or burst, causing an intracranial hemorrhage, a potentially fatal disorder. One of the critical steps in brain stroke imaging is the categorization of intracranial hemorrhages. This article aims to perform texture-based intracranial hemorrhage computed tomography image classification with ensemble and machine learning classifiers using local binary pattern, local ternary pattern, and Weber local descriptor approach. The most common texture feature extraction method is a local binary pattern; however, other methods such as local ternary pattern, Weber local descriptor, and local binary pattern are also utilized for texture feature extraction. For all the extracted texture codes, histograms are applied, and finally, single feature vector texture codes by performing the concatenation of the bin. The efficacy of feature descriptors for classifying images into abnormal and normal classes was assessed using ten distinct classifiers. The result shows that Weber's local descriptor shows promising results in texture code classification for ensemble-based classification. For all texture-based methods, classification accuracy was better than standard machine learning. According to the findings of the experiment, the Random Forest classifier outperformed all other classifiers in terms of classification accuracy (86.55%), recall (86.31%), precision (87.23%), sensitivity (86.31%), specificity (86.81%), and F1-score (86.77%) for the Weber local descriptor.

Enhanced accuracy for sms spam detection using One Dimensional Ternary Patterns (1D-TP) and firefly algorithm

Enhanced accuracy for sms spam detection using One Dimensional Ternary Patterns (1D-TP) and firefly algorithm

Local-Ternary-Pattern-Based Associated Histogram Equalization Technique for Cervical Cancer Detection.

Every year, cervical cancer is a leading cause of mortality in women all over the world. This cancer can be cured if it is detected early and patients are treated promptly. This study proposes a new strategy for the detection of cervical cancer using cervigram pictures. The associated histogram equalization (AHE) technique is used to improve the edges of the cervical image, and then the finite ridgelet transform is used to generate a multi-resolution picture. Then, from this converted multi-resolution cervical picture, features such as ridgelets, gray-level run-length matrices, moment invariant, and enhanced local ternary pattern are retrieved. A feed-forward backward propagation neural network is used to train and test these extracted features in order to classify the cervical images as normal or abnormal. To detect and segment cancer regions, morphological procedures are applied to the abnormal cervical images. The cervical cancer detection system's performance metrics include 98.11% sensitivity, 98.97% specificity, 99.19% accuracy, a PPV of 98.88%, an NPV of 91.91%, an LPR of 141.02%, an LNR of 0.0836, 98.13% precision, 97.15% FPs, and 90.89% FNs. The simulation outcomes show that the proposed method is better at detecting and segmenting cervical cancer than the traditional methods.

Fine-tuning convolutional neural network based on relaxed Bayesian-optimized support vector machine for random-valued impulse noise removal

Denoising convolutional neural network (DnCNN) has achieved competitive denoising performance for Gaussian noise using residual learning. The same idea can also be applied in the impulse noise (IN) removal by changing the training data. However, as a discriminative learning-based algorithm, DnCNN lacks flexibility in removing spatially varying noise, such as random-valued impulse noise (RVIN). To address this problem, we propose a fine-tuning CNN based on relaxed Bayesian-optimized support vector machine (BO-SVM) for RVIN removal using a relaxed BO-SVM as a detector and a fine-tuning DnCNN as a denoiser. The relaxed BO-SVM is developed to classify all pixels of the test image into two categories: noise and noise-free. A relaxed rule is provided to preserve the information of slightly corrupted pixels, and the system is trained with an optimal feature set, which comprises of newly introduced local ternary pattern with previously used gray-level difference, average background difference, and median value difference. The key hyperparameters of the relaxed SVM are obtained using the Bayesian optimization technique. Then the DnCNN works for image reconstruction through fine-tuning experiments by the use of the corrupted pixels identified by the detector. The experimental results show that the proposed method reports promising results on RVIN removal over other state-of-the-art denoising methods.

Enhanced three‐dimensional model reconstruction based on local ternary pattern‐guided fusion of multi‐exposure images

AbstractComputer vision applications usually rely on the features extracted from input images with good visibility. Image acquisition systems may produce degraded images with low contrast or distorted colours. For instance, bad weather (haze, fog) can cause images captured outdoor with low visibility. Image processing algorithms generally assume that the input image is the scene radiance. Haze removal, with the recovery of image radiance, ensures reliable features extracted from images and the image processing algorithm can achieve optimal performance. Inspired by the concept of image dehazing, the authors propose an image enhancement method that can be used to improve the visibility of the images. Each original image is first transformed into multiple exposure images by means of gamma‐correction operations and adaptive histogram equalization. The transformed images are analyzed by the computation of the local ternary pattern. The image is then enhanced, with each pixel generated from the set of transformed image pixels weighted by a function of the local pattern feature. The authors evaluate their proposed method on four benchmark image dehazing datasets. The quantitative results show that our method outperforms many deterministic algorithms and deep learning models. Moreover, the authors investigate the impact of image enhancement on a practical image‐based application—the reconstruction of three‐dimensional (3D) model of survey scene. Accurate 3D model reconstruction depends on high‐quality images. Degraded images will result in large errors in the reconstructed 3D model. Experimentations have been carried out on outdoor and indoor surveys. Our analysis finds that when fed into the photogrammetry software, the images enhanced by the authors’ method can reconstruct 3D scene models with sub‐millimetre mean errors, which are much better than those with the original images. As shown in the visual and quantitative results of 3D model reconstruction, the authors’ proposed method also outperforms other image enhancement methods.

Surface curvature based completed local ternary pattern for texture image classification

Surface curvature based completed local ternary pattern for texture image classification

Locating robust patterns based on invariant of LTP-based features

• A novel concept of robust patterns (called RLTP) is introduced by considering the invariance of LTP-based features. • A crucial derivative of CLTP (called RCLTP) is proposed to be in accordance with RLTP. • An efficient application is presented to take into account RCLTP for local DT representation. • Local optical-flow-based features are taken into account DT motions for boosting the discrimination. • Just addressing the robust patterns, our RCLTP descriptor obtains very good performance compared to state of the art. Efficiently representing Dynamic Textures (DTs) based on salient features is one of the considerable challenges in computer vision. Locating these features can be obstructed due to the impact of encoding factors. In this article, a novel concept of Robust Local Ternary Patterns (RLTP) is introduced in consideration of the invariance of Local Ternary Patterns (LTP) subject to the deviation of thresholds. Our locating process is able to simultaneously encapsulate the discrimination of local features, and deal with the noise sensibility caused by a small gray-scale change of local neighbors. RLTP is then adapted to the completed LTP model to form an efficient operator for capturing completely properties of RLTP. Finally, RLTP is taken into account for the DT description, where the robust patterns of spatial-temporal features and optical-flow-based motions are exploited to improve the performance. Experiments have clearly corroborated the efficacy of our proposed approach.

Enhancing the Classification Accuracy of Rice Varieties by Using Convolutional Neural Networks

The aim of this study is to enhance the classification accuracy of rice varieties that are quite similar in external observation. In this study, 17 rice grain varieties popularly planted in Vietnam are classified with an Artificial Neural Network (ANN) and Convolutional Neural Network (CNN) models. The two CNN models (modified VGG16 and modified ResNet50) are based on pre-trained VGG16 and Resnet50 models. Two datasets are used in the experiments: a feature dataset extracted using an extended improved local ternary pattern (extended ILTP) method, and an image dataset generated with a data augmentation technique. The feature dataset was fed into the ANN, while the image dataset was fed into the CNN models. The highest classification accuracies of ANN, modified VGG16, and modified ResNet50 models are 92.82%, 96.41%, and 97.88%, respectively. The results show that the modified VGG16 and ResNet50 models significantly improved classification accuracy of the 17 varieties of rice. In addition, the experiments show that the dimensions of the image dataset can affect the performance of the CNN models. This research can be developed for applications of rice varieties classification and identification.

Surface curvature based Completed Local Ternary Pattern for Texture Image Classification

Surface curvature based Completed Local Ternary Pattern for Texture Image Classification