Local Directional Pattern Research Articles

Accuracy Assessment & Classification of Keratosis Skin Lesion Images using Feature Extraction & Classification Algorithms-LBP, LDP& HOG

There are hundreds of human-affected skin diseases. The most severe skin disorders may have identical symptoms, so recognizing the distinctions between them is crucial. People should work closely with a dermatologist to identify and manage every skin disorder and insure it does not impact their lifestyle.Actinic keratosis (AK), that is also classified as solar or senile keratosis; is a pre-malignant crusty, thick skin area. It is a disorder of epidermal keratinocytes, induced by UV radiation upon the skin. While pre-cancerous in nature, they can develop into a form of skin cancer called carcinoma if left unaddressed. The other type of keratosis dealt within this paper is seborrheic keratosis, which are brown or black, thick, wart-like, waxy oval-shaped, slightly raised skin surfaces. The growths aren't damaging. Nevertheless, in some instances it can be impossible to differentiate a seborrheic keratosis from melanoma, which is a very dangerous form of skin cancer.Nevus (or moles) skin lesions are ones which are benign, where it may very rarely turn into melanoma skin cancer. In this article, along with techniques for extracting features (LDP [Local Directional Patterns], LBP [Local Binary Patterns] and HOG [Histogram of Oriented Gradients]),we have used an SVM classifier for the classification of Keratosis and also nevus skin photos. The LBP, LDP and HOG are means to extract features; these images are subsequently used for identification of derived features from these methods or algorithms and classified by the SVM (Support Vector Machine) classifier. For many of the classifications of keratosis and nevus skin images using these algorithms, we have obtained accuracy nearly above 80 %, whereby the LBP system together with the SVM classifier was the most powerful attribute extraction tool of the three with their polynomial kernel type. Using this algorithm-classifier,the main AK and nevus skin lesion images can be detected and diagnosed by the doctors in its early stage itself,thus helping save lives.

Retrieval of colour and texture images using local directional peak valley binary pattern

Many content-based image retrieval (CBIR) methods are being developed to store more and more information about images in shorter feature vectors and to improve image retrieval rate. In the proposed method, two-step approach to CBIR has been developed. The first step generates an image mask from local binary pattern (LBP). This LBP mask is then utilized to draw comparison between the centre pixel and the eight surrounding pixels. The second step involves drawing the peak and valley patterns of local directional binary pattern for each image which is then combined with the colour histogram to retrieve similar images. Existing methods suffer from lower average image retrieval accuracy even with larger feature vectors. The proposed method overcomes such problems through shorter feature vectors that can store more information about the image. As illustrated through experimental results, the proposed method produces promising results with shorter feature vector of length 56 and improved image retrieval rate of about 5–10%. Our method outperforms similar techniques when tested with public data sets.

RETRACTED ARTICLE: Local Directional Maximum Edge Patterns for facial expression recognition

Cognitive science and neuroscience use human facial expressions of emotion. Every single facial expression can be seen at different passions in a face space. Nowadays, facial expression recognition and analysis is vital due to the demand of introducing advanced biometric applications in every domain space. The imperative task in facial expressions of emotion classification is precise feature extraction, which helps to get detailed description of facial marks. Existing feature descriptors are suffering from various problems such as intensity variations, discrimination, vulnerability etc. In this paper, propose a new feature descriptor method called LDMEP (Local Directional Maximum Edge Patterns) for facial expression analysis to overcome the hindrance. We calculated the gradients in four directions of reference pixel to elicit the more feature for better recognition instead of calculating the local differences among neighboring pixels. We also access the orientations of the pixels then thresholded based on the dynamic threshold to avoid the featureless area calculation. Furthermore, we considered only dominant magnitude and orientation directions instead of all eight directions to generate feature. Thus, imperative and efficient features are covered in dominant positions to detect the strong edges. The paper confers that how the subsequent model can be used for the recognition of facial expression of emotion.

Kinship verification through facial images using multiscale and multilevel handcrafted features

We address kinship verification, which is a challenging problem in computer vision and pattern discovery. It has several applications, such as organizing photoalbums, recognizing resemblances among humans, and finding missing children. We present a system for facial kinship verification based on several kinds of texture descriptors (local binary patterns, local ternary patterns, local directional patterns, local phase quantization, and binarized statistical image features) with pyramid multilevel (PML) face representation for feature extraction along with our proposed paired feature representation and our proposed robust feature selection to reduce the number of features. The proposed approach consists of the following three main stages: (1) face preprocessing, (2) feature extraction and selection, and (3) kinship verification. Extensive experiments are conducted on five publicly available databases (Cornell, UB KinFace, Family 101, KinFace W-I, and KinFace W-II). Additionally, we provided a wide experiment for each stage to find the best and most suitable settings. We present many comparisons with state-of-the-art methods and through these comparisons, it appears that our experiments show stable and good results.

Analysis of Melanoma Lesion Images using Feature Extraction & Classification Algorithms

Among the most dangerous of cancers found in human beings, skin cancer is the prevalent one. These are of various forms. The most sporadic among them is melanoma. Early phase identification of melanoma will be helpful in curing it. Intensive skin exposure to UV radiation is the principal cause of melanoma. In this article, along with other techniques for extracting features (LDP [Local Directional Patterns], LBP [Local Binary Patterns], Convolutional Neural Networks [CNN]), we have used an SVM classifier for the study of melanoma skin photos. Such suggested algorithms are best graded when opposed to other recognition schemes. The LBP and LDP gives us means to extract features; these figures are subsequently used for identification of derived features from these methods or algorithms and classified or separated by the SVM (Support Vector Machine) classifier. For many of the classifications of melanoma skin images using these algorithms, we have accuracy nearly above 80 %, whereby the LBP system together with the SVM classifier was the most powerful attribute extraction tool of the three with their polynomial kernel type. Thus using this algorithm-classifier, the melanoma skin lesion images can be detected and diagnosed by the doctors in its early stage itself, resultantly, helping save lives.

A Conceptual Model for Automated Attendance Marking System using Facial Recognition

The project represents an automated attendance system based on face recognition using discriminative local binary pattern and local directional pattern descriptors. The proposed system involves face detection, Features extraction and matching. The face detection is to detect faces based on Viola Jones algorithm . In feature extraction stage, the discriminative local binary pattern is used for different object texture feature extraction process. The proposed method with new features retain the contrast information of image patterns. The Facial recognition (or face recognition) is a type of biometric application that can identify a specific individual in a digital image by analyzing and comparing patterns based on the data stored in database. Smart attendance is a real time face recognition used for handling day to day activities of the employees and students. In manual attendance system there are several issues like fake attendance and mistakenly marked absent by carelessness of the faculty/teachers/lectures. In order to overcome these problems we are using this smart attendance system. Here multiple faces are detected and recognized with trained with various features. The automated face attendance marking system gives accurate performance.

Local transform directional pattern and optimization driven DBN for age estimation

Age estimation is an interesting and challenging research area, gaining significant importance in the recent era and is employed in various applications, such as intelligent surveillance, face recognition, biometrics, and so on. Various techniques are employed in the literature for age estimation from the face images. This paper introduces the age estimation scheme from the face image, and estimation is done by defining a novel feature extraction strategy, named Local Transform Directional Pattern (LTDP). The database containing the input images has many unwanted regions, and thus, the Viola Jones algorithm detects the required face region. After detecting the face regions, active appearance model extracts the active appearance features, and the proposed LTDP extracts the texture features. The proposed LTDP feature extraction model modifies the existing Local Directional Pattern (LDP) with several other texture feature extraction models. After the feature extraction, the Cuckoo search based Deep Belief Network (CDBN) classifier estimates the age of the person from the face image based on the extracted features. The simulation results reveal that the proposed LTDP with the CDBN classifier achieved high performance with the values of 2.3416, 0.9803, and 0.9724 for MAE, AEO, and AEM, respectively.

Self‐adaptive weighted synthesised local directional pattern integrating with sparse autoencoder for expression recognition based on improved multiple kernel learning strategy

This study presents a novel method for solving facial expression recognition (FER) tasks which uses a self‐adaptive weighted synthesised local directional pattern (SW‐SLDP) descriptor integrating sparse autoencoder (SA) features based on improved multiple kernel learning (IMKL) strategy. The authors’ work includes three parts. Firstly, the authors propose a novel SW‐SLDP feature descriptor which divides the facial images into patches and extracts sub‐block features synthetically according to both distribution information and directional intensity contrast. Then self‐adaptive weights are assigned to each sub‐block feature according to the projection error between the expressional image and neutral image of each patch, which can highlight such areas containing more expressional texture information. Secondly, to extract a discriminative high‐level feature, they introduce SA for feature representation, which extracts the hidden layer representation including more comprehensive information. Finally, to combine the above two kinds of features, an IMKL strategy is developed by effectively integrating both soft margin learning and intrinsic local constraints, which is robust to noisy condition and thus improve the classification performance. Extensive experimental results indicate their model can achieve competitive or even better performance with existing representative FER methods.

Person-Independent Facial Expression Recognition Based on Improved Local Binary Pattern and Higher-Order Singular Value Decomposition

The recognition rate of person-independent facial expression is generally not high, which limits the practical application of facial expression recognition. Aiming at this problem, this paper analyzes the reasons for the low recognition rate of person-independent facial expression, and proposes a recognition algorithm of person-independent facial expression based on improved LBP (Local Binary Pattern) and HOSVD (Higher-Order Singular Value Decomposition). The algorithm has the following contributions of facial expression recognition framework. In the stage of facial expression feature extraction, the transient features extracted by LDP(Local Directional Pattern) and the persistent features extracted by CBP(Centralized Binary Pattern) are integrated to improve the discrimination of facial expression features. Moreover, in the stage of facial expression classification and recognition, the traditional nearest neighbor classification is changed into k-nearest neighbor pre-classification, and the regional energy calculated by HOSVD is used to determine the similarity of two images for secondary classification. Finally, in the extended Cohn-Kanade dataset and Oulu-CASIA NIR&VIS facial expression database, the theoretical analysis and experimental results show that the method has better recognition effect for solving the problem of person-independent facial expression recognition.

Leaf Recognition Based on Elliptical Half Gabor and Maximum Gap Local Line Direction Pattern

Plant identification via leaf images is very meaningful to agricultural information. The existing methods were based on one or two kinds of the three distinct characteristics in leaf images including leaf contours, textures and veins. This limits their recognition performance and scope of application. This paper describes a novel counting-based leaf recognition method, which can directly and effectively combine all of the three kinds of significant characteristics in leaf images. In order to obtain the stable and independent local line responses from leaf contour, texture and vein, elliptical half Gabor is introduced and convoluted with the raw grayscale leaf images, and then maximum gap local line direction patterns are extracted from the local line responses and normalized in direction by cyclically right shifting these patterns until the most numerous bit plane with a value of 1 to the left bit. The histogram of the normalized patterns is calculated and regarded as the counting-based local structure descriptor, and support vector machine is utilized as the classifier. Experimental results on three frequently used leaf databases show that the proposed approach yields a better performance in terms of the classification accuracy, applicability and feasibility in comparison with the state of the art methods.

LDOP: local directional order pattern for robust face retrieval

The local descriptors have gained wide range of attention due to their enhanced discriminative abilities. It has been proved that the consideration of multi-scale local neighborhood improves the performance of the descriptor, though at the cost of increased dimension. This paper proposes a novel method to construct a local descriptor using multi-scale neighborhood by finding the local directional order among the intensity values at different scales in a particular direction. Local directional order is the multi-radius relationship factor in a particular direction. The proposed local directional order pattern (LDOP) for a particular pixel is computed by finding the relationship between the center pixel and local directional order indexes. It is required to transform the center value into the range of neighboring orders. Finally, the histogram of LDOP is computed over whole image to construct the descriptor. In contrast to the state-of-the-art descriptors, the dimension of the proposed descriptor does not depend upon the number of neighbors involved to compute the order; it only depends upon the number of directions. The introduced descriptor is evaluated over the image retrieval framework and compared with the state-of-the-art descriptors over challenging face databases such as PaSC, LFW, PubFig, FERET, AR, AT&T, and ExtendedYale. The experimental results confirm the superiority and robustness of the LDOP descriptor.

Biomedical image retrieval by using local directional edge binary patterns and Zernike moments

This paper presents an efficient approach with the reduced length of feature vector for biomedical image retrieval by using global and local features of an image. In order to extract the local features, a new algorithm, local directional edge binary pattern (LDEBP) has been designed. It gathers information from all the possible directions, i.e., 00, 450, 900 and 1350 for every pixel in the image. The directional information is calculated based on the sign code magnitudes of local differences from the center pixel to its directional pixels. For every pixel, four edges will be calculated by using all the directional information. Lower order Zernike moments are used for extracting the global and shape features of an image. The combination of shape and texture descriptors for biomedical image retrieval showed significant results compared to the state of the art algorithms like LDEP, ZM and LBDP on benchmark database like Emphysema-CT and OASIS-MRI.

Direction-invariant binary pattern-encoded descriptor for texture classification and retrieval

Texture feature captures the structural arrangement details of pixels present in a given image. The local binary pattern (LBP) and its modified versions have been used to extract discriminant information about the structural arrangement. A unique textural representation obtained from LBP and its modified versions are affected by rotating the structural arrangement of the pixels available in a particular region. The weight of the binary values obtained from the LBP and its modified versions are constant. The structural arrangement of the local region is effectively determined from the direction-oriented edge information [i.e., local directional pattern (LDP)]. The LDP is also nonrobust toward the structural arrangement changes in the local region. We give a direction-invariant binary pattern to represent the texture details available in a particular location by encoding edge response along with the LBP of that region. The direction-invariant property is derived from the weight of the edge response available in a particular region. The edge response around each local region of the image is captured using the compass edge detector. It detects the edge details from eight directions. The process of encoding eight direction edge responses along with a binary pattern obtained from the center pixel difference gives a more robust texture descriptor than the other rotation-invariant descriptors used in texture feature extraction. The proposed descriptor is tested over five standard databases, namely Brodatz, color Brodatz, KTH-TIPS2b, FTVL, and free photo. The performance of the proposed rotation-invariant texture feature is more effective than the existing rotation-invariant local texture feature representations involved in the image classification and retrieval experiments.

A Secure Boot Loader System for Loading an Operating System with Fingerprint Authentication

In today’s world, people are constantly on move and portable systems are in demand. With technology advancement, people exploit different types of memory devices for a portable system. For any external boot medium, the BIOS boot order setting change is required. The dynamic boot loader successfully eliminated this dependency and allowed the user to directly boot from any portable USB. The usage of USB has grown exponentially in recent years and securing it has become a major concern. In this paper, the USB is devised as a highly secured portable boot medium with fingerprint authentication to ensure data security. It performs feature extraction by combining both Local Directional Pattern (LDP) and Histograms of Oriented Gradients (HOG) which improves the accuracy rate. The classification is performed by random forest classifier, such that the intended users alone are granted access to the private storage area of the USB drive.

An effective image retrieval based on optimized genetic algorithm utilized a novel SVM-based convolutional neural network classifier

Image retrieval is the process of retrieving images from a database. Certain algorithms have been used for traditional image retrieval. However, such retrieval involves certain limitations, such as manual image annotation, ineffective feature extraction, inability capability to handle complex queries, increased time required, and production of less accurate results. To overcome these issues, an effective image retrieval method is proposed in this study. This work intends to effectively retrieve images using a best feature extraction process. In the preprocessing of this study, a Gaussian filtering technique is used to remove the unwanted data present in the dataset. After preprocessing, feature extraction is applied to extract features, such as texture and color. Here, the texture feature is categorized as a gray level cooccurrence matrix, whereas the novel statistical and color features are considered image intensity-based color features. These features are clustered by k-means clustering for label formation. A modified genetic algorithm is used to optimize the features, and these features are classified using a novel SVM-based convolutional neural network (NSVMBCNN). Then, the performance is evaluated in terms of sensitivity, specificity, precision, recall, retrieval and recognition rate. The proposed feature extraction and modified genetic algorithm-based optimization technique outperforms existing techniques in experiments, with four different datasets used to test the proposed model. The performance of the proposed method is also better than those of the existing (RVM) regression vector machine, DSCOP, as well as the local directional order pattern (LDOP) and color co-occurrence feature + bit pattern feature (CCF + BPF) methods, in terms of the precision, recall, accuracy, sensitivity and specificity of the NSVMBCNN.

Mutual Information based hybrid model and deep learning for Acute Lymphocytic Leukemia detection in single cell blood smear images

Background and objectiveDue to the development in digital microscopic imaging, image processing and classification has become an interesting area for diagnostic research. Various techniques are available in the literature for the detection of Acute Lymphocytic Leukemia from the single cell blood smear images. The purpose of this work is to develop an effective method for leukemia detection. MethodsThis work has developed deep learning based leukemia detection module from the blood smear images. Here, the detection scheme carries out pre-processing, segmentation, feature extraction and classification. The segmentation is done by the proposed Mutual Information (MI) based hybrid model, which combines the segmentation results of the active contour model and fuzzy C means algorithm. Then, from the segmented images, the statistical and the Local Directional Pattern (LDP) features are extracted and provided to the proposed Chronological Sine Cosine Algorithm (SCA) based Deep CNN classifier for the classification. ResultsFor the experimentation, the blood smear images are considered from the AA-IDB2 database and evaluated based on metrics, such as True Positive Rate (TPR), True Negative Rate (TNR), and accuracy. Simulation results reveal that the proposed Chronological SCA based Deep CNN classifier has the accuracy of 98.7%. ConclusionsThe performance of the proposed Chronological SCA-based Deep CNN classifier is compared with the state-of-the-art methods. The analysis shows that the proposed classifier has comparatively improved performance and determines the leukemia from the blood smear images.

Image descriptor based on local color directional quaternionic pattern

Most of the existing local binary pattern methods discard local color differences by holding their binary information. And local spatial information is also neglected. To address these problems, a robust color image descriptor local color directional quaternionic pattern (LCDQP) is proposed. In the descriptor, the color distance map (CDM) is generated in the RGB color space to capture the color distribution among three channels. According to the distribution of CDM elements and their neighbors, four edge models are defined to describe the change trend of the original image. Then, based on this, the LCDQP strings are gained according to the distribution of four models in the four directions of 0 deg, 45 deg, 90 deg, and 135 deg. Finally, an effective quaternionic code method is adopted to construct the LCDQP descriptor. The proposed descriptor not only captures the local color features but also reflects the spatial structure information. Experiments on four representative databases demonstrate that the proposed descriptor is superior to other state-of-the-art approaches.

Local directional relation pattern for unconstrained and robust face retrieval

Face recognition is still a very demanding area of research. This problem becomes more challenging in unconstrained environment and in the presence of several variations like pose, illumination, expression, etc. Local descriptors are widely used for this task. The most of the existing local descriptors consider only few immediate local neighbors and not able to utilize the wider local information to make the descriptor more discriminative. The wider local information based descriptors mainly suffer due to the increased dimensionality. In this paper, this problem is solved by encoding the relationship among directional neighbors in an efficient manner. The relationship between the center pixel and the encoded directional neighbors is utilized further to form the proposed local directional relation pattern (LDRP). The descriptor is inherently uniform illumination invariant. The multi-scale mechanism is also adapted to further boost the discriminative ability of the descriptor. The proposed descriptor is evaluated under the image retrieval framework over face databases. Very challenging databases like PaSC, LFW, PubFig, ESSEX, FERET, AT&T, and FaceScrub are used to test the discriminative ability and robustness of LDRP descriptor. Results are also compared with the recent state-of-the-art face descriptors such as LBP, LTP, LDP, LDN, LVP, DCP, LDGP and LGHP. Very promising performance is observed using the proposed descriptor over very appealing face databases as compared to the existing face descriptors. The proposed LDRP descriptor also outperforms the pre-trained ImageNet CNN models over large-scale FaceScrub face dataset. Moreover, it also outperforms the deep learning based DLib face descriptor in many scenarios.

Local Prominent Directional Pattern을 이용한 얼굴 사진과 스케치 영상 성별인식 방법

본 논문에서는 성별 인식을 위해 얼굴 영상을 효과적으로 기술하는 새로운 지역 패턴 방법 Local Prominent Directional Pattern (LPDP)를 제안한다. 제안된 LPDP 방법은 성별 인식에 중요한 얼굴 모양을 명확하게 구분하기 위해 주변 패턴이 누 적된 히스토그램을 통계적으로 분석하고 패턴 변화가 크게 발생하는 픽셀을 부호화 한다. 통계적인 정보를 사용하는 얼굴 모 양 구분에 중요한 뚜렷한 에지 방향 패턴 영역을 구분하는 중요한 정보를 제공 할 수 있다. 이는 뚜렷한 에지 방향 패턴이 나 타나는 영역의 주변도 유사한 에지 방향 패턴이 나타내기 때문에 통계적으로 특정 방향이 히스토그램에 많이 누적될 수 있기 때문이다. 또한 통계적인 방법은 주변 영역의 정보를 많이 수용하기 때문에 잡음으로 발생하는 에지 방향 변화 오류에 강력한 장점이 있다. 제안된 방법은 기존 방법들 보다 더 강력한 성별인식에 중요한 얼굴 모양 구분 능력을 보여주면서 국소적으로 발생하는 잡음에 견고함을 보여준다. 우리는 제안된 방법의 성능을 평가하기 위해 밝기, 표정, 연령, 머리 포즈가 변화하는 성 별 인식 데이터 셋에 다양한 실험을 실험 했고 기존 방법 보다 제안된 방법의 성능이 우수함을 입증했다.

Learner posture recognition via a fusing model based on improved SILTP and LDP

The learner’s body posture reflects the learner’s learning state. The learner’s posture recognition can effectively evaluate their learning state, which plays an important role in the teacher’s teaching process. In this paper, a new method for learner’s gesture recognition is proposed, which fuses the improved scale invariant local ternary pattern (SILTP) and the local directional pattern (LDP). Firstly, a multi-scale weighted adaptive SILTP (MWA-SILTP) algorithm is proposed. The dynamic threshold of the current neighborhood is adaptively generated according to the dispersion degree of contrast values in global and local neighborhoods, and SILTP coding is carried out to obtain adaptive SILTP. And the concept of multi-scale is introduced. By changing the sampling radius, the adaptive SILTPs of different scales are obtained. The adaptive SILTPs of different scales are merged with different weights to represent the image in multi-resolution. The MWA-SILTP algorithm is used to extract the feature of the learner’s posture image. Secondly, the LDP algorithm is used to extract the feature of the learner’s posture image. Finally, the two features are merged, and the support vector machine is used for classification and recognition. The improved SILTP can get more feature information and has stronger adaptability. The LDP algorithm has advantages in anti-interference and can extract edge information better. The fusing model of this paper fully utilizes the advantages of the two algorithms. Experimental results show that the proposed method can effectively recognize learner’s posture of sitting, raising hand and lowering head.