Edge Detection and Contour Based Ear Recognition Scheme

2019 ◽  
Vol 12 (3) ◽  
pp. 74-92
Author(s):  
Deven Trivedi ◽  
Rohit Thanki ◽  
Surekha Borra
Keyword(s):  
Pattern Recognition ◽  
Histogram Equalization ◽  
Biometric Data ◽  
Snake Model ◽  
Ear Recognition ◽  
Computer Vision Technology ◽  
Human Ear ◽  
Edge Features ◽  
Matching Score ◽  
Better Than

In recent days, with the advancements in computer vision technology pattern recognition for biometric data has been the focus of many researchers. The human ear can be used to assist in the recognition of an individual. In this article, a new scheme for ear recognition is presented, based on edge features such as the helix shape and contours between the edge pixels. First, an ear image is detected from the acquired image using a snake model-based image segmentation technique, and then histogram equalization is applied to form an enhanced ear image. After that, an Infinite Symmetric Exponential Filter (ISEF) edge is applied to the image, the contouring of edges is calculated, and then the contour values of pixels are extracted as ear features. Finally, the ear matching is performed between query ear features and enrolled ear features. Based on the matching score, the decision about individual authentication is performed. The experimental results showed that this proposed scheme performs better than existing schemes in the literature.

Combining Block DCV and Support Vector Machine for Ear Recognition

2016 ◽  
Vol 8 (2) ◽  
pp. 36-44 ◽  
Author(s):  
Zhao Hailong ◽  
Yi Junyan
Keyword(s):  
Support Vector Machine ◽  
Feature Extraction ◽  
Image Retrieval ◽  
Construction Method ◽  
Experimental Results ◽  
Support Vector ◽  
Recognition Method ◽  
Ear Recognition ◽  
Human Ear ◽  
Better Than

In recent years, automatic ear recognition has become a popular research. Effective feature extraction is one of the most important steps in Content-based ear image retrieval applications. In this paper, the authors proposed a new vectors construction method for ear retrieval based on Block Discriminative Common Vector. According to this method, the ear image is divided into 16 blocks firstly and the features are extracted by applying DCV to the sub-images. Furthermore, Support Vector Machine is used as classifier to make decision. The experimental results show that the proposed method performs better than classical PCA+LDA, so it is an effective human ear recognition method.

Combining Block DCV and Support Vector Machine for Ear Recognition

2018 ◽  
pp. 774-783
Author(s):  
Zhao Hailong ◽  
Yi Junyan
Keyword(s):  
Support Vector Machine ◽  
Feature Extraction ◽  
Image Retrieval ◽  
Construction Method ◽  
Experimental Results ◽  
Support Vector ◽  
Recognition Method ◽  
Ear Recognition ◽  
Human Ear ◽  
Better Than

In recent years, automatic ear recognition has become a popular research. Effective feature extraction is one of the most important steps in Content-based ear image retrieval applications. In this paper, the authors proposed a new vectors construction method for ear retrieval based on Block Discriminative Common Vector. According to this method, the ear image is divided into 16 blocks firstly and the features are extracted by applying DCV to the sub-images. Furthermore, Support Vector Machine is used as classifier to make decision. The experimental results show that the proposed method performs better than classical PCA+LDA, so it is an effective human ear recognition method.

Pattern-Recognition: A Comparison of the Performance of Clinicians and Non-Clinicians - with a Note on the Performance of a Computer-Based System

1972 ◽  
Vol 11 (01) ◽  
pp. 32-37 ◽  
Author(s):  
F. T. DE DOMBAL ◽  
J. C. HORROCKS ◽  
J. R. STANILAND ◽  
P. J. GUILLOU
Keyword(s):  
Pattern Recognition ◽  
Computer System ◽  
Computer Aided Diagnosis ◽  
Clinical Interview ◽  
Computer Performance ◽  
Pattern Size ◽  
Computer Aided ◽  
Computer Based ◽  
Aided Diagnosis ◽  
Better Than

This paper describes a series of 10,500 attempts at »pattern-recognition« by two groups of humans and a computer based system. There was little difference between the performances of 11 clinicians and 11 other persons of comparable intellectual capability. Both groups’ performances were related to the pattern-size, the accuracy diminishing rapidly as the patterns grew larger. By contrast the computer system increased its accuracy as the patterns increased in size.It is suggested (a) that clinicians are very little better than others at pattem-recognition, (b) that the clinician is incapable of analysing on a probabilistic basis the data he collects during a traditional clinical interview and examination and (c) that the study emphasises once again a major difference between human and computer performance. The implications as - regards human- and computer-aided diagnosis are discussed.

Human Ear Pattern Recognition System

2017 ◽  
Vol 7 (8) ◽  
pp. 117 ◽  
Author(s):  
V. Jagan Naveen ◽  
K. Krishna Kishore ◽  
P. Rajesh Kumar
Keyword(s):  
Pattern Recognition ◽  
Expectation Maximization Algorithm ◽  
Recognition System ◽  
Modern World ◽  
Person Identification ◽  
Biometric Systems ◽  
Different Types ◽  
Recognition Systems ◽  
Human Ear ◽  
Pca Algorithm

In the modern world, human recognition systems play an important role to   improve security by reducing chances of evasion. Human ear is used for person identification .In the Empirical study on research on human ear, 10000 images are taken to find the uniqueness of the ear. Ear based system is one of the few biometric systems which can provides stable characteristics over the age. In this paper, ear images are taken from mathematical analysis of images (AMI) ear data base and the analysis is done on ear pattern recognition based on the Expectation maximization algorithm and k means algorithm.  Pattern of ears affected with different types of noises are recognized based on Principle component analysis (PCA) algorithm.

Width-based distance measures on interval-valued intuitionistic fuzzy sets

2021 ◽  
pp. 1-13
Author(s):  
Xi Li ◽  
Chunfeng Suo ◽  
Yongming Li
Keyword(s):  
Pattern Recognition ◽  
Fuzzy Sets ◽  
Medical Diagnosis ◽  
Intuitionistic Fuzzy Sets ◽  
Distance Measures ◽  
Intuitionistic Fuzzy ◽  
Interval Valued ◽  
Dissimilarity Function ◽  
Better Than

An essential topic of interval-valued intuitionistic fuzzy sets(IVIFSs) is distance measures. In this paper, we introduce a new kind of distance measures on IVIFSs. The novelty of our method lies in that we consider the width of intervals so that the uncertainty of outputs is strongly associated with the uncertainty of inputs. In addition, better than the distance measures given by predecessors, we define a new quaternary function on IVIFSs to construct the above-mentioned distance measures, which called interval-valued intuitionistic fuzzy dissimilarity function. Two specific methods for building the quaternary functions are proposed. Moreover, we also analyzed the degradation of the distance measures in this paper, and show that our measures can perfectly cover the measures on a simpler set. Finally, we provide illustrative examples in pattern recognition and medical diagnosis problems to confirm the effectiveness and advantages of the proposed distance measures.

scholarly journals Deep learning model for classifying endometrial lesions

2021 ◽  
Vol 19 (1) ◽  
Author(s):  
YunZheng Zhang ◽  
ZiHao Wang ◽  
Jin Zhang ◽  
CuiCui Wang ◽  
YuShan Wang ◽  
...  
Keyword(s):  
Endometrial Cancer ◽  
Endometrial Hyperplasia ◽  
Histogram Equalization ◽  
Atypical Hyperplasia ◽  
Endometrial Polyps ◽  
Submucous Myoma ◽  
Objective Model ◽  
Medical University ◽  
Deep Learning Model ◽  
Better Than

Abstract Background Hysteroscopy is a commonly used technique for diagnosing endometrial lesions. It is essential to develop an objective model to aid clinicians in lesion diagnosis, as each type of lesion has a distinct treatment, and judgments of hysteroscopists are relatively subjective. This study constructs a convolutional neural network model that can automatically classify endometrial lesions using hysteroscopic images as input. Methods All histopathologically confirmed endometrial lesion images were obtained from the Shengjing Hospital of China Medical University, including endometrial hyperplasia without atypia, atypical hyperplasia, endometrial cancer, endometrial polyps, and submucous myomas. The study included 1851 images from 454 patients. After the images were preprocessed (histogram equalization, addition of noise, rotations, and flips), a training set of 6478 images was input into a tuned VGGNet-16 model; 250 images were used as the test set to evaluate the model’s performance. Thereafter, we compared the model’s results with the diagnosis of gynecologists. Results The overall accuracy of the VGGNet-16 model in classifying endometrial lesions is 80.8%. Its sensitivity to endometrial hyperplasia without atypia, atypical hyperplasia, endometrial cancer, endometrial polyp, and submucous myoma is 84.0%, 68.0%, 78.0%, 94.0%, and 80.0%, respectively; for these diagnoses, the model’s specificity is 92.5%, 95.5%, 96.5%, 95.0%, and 96.5%, respectively. When classifying lesions as benign or as premalignant/malignant, the VGGNet-16 model’s accuracy, sensitivity, and specificity are 90.8%, 83.0%, and 96.0%, respectively. The diagnostic performance of the VGGNet-16 model is slightly better than that of the three gynecologists in both classification tasks. With the aid of the model, the overall accuracy of the diagnosis of endometrial lesions by gynecologists can be improved. Conclusions The VGGNet-16 model performs well in classifying endometrial lesions from hysteroscopic images and can provide objective diagnostic evidence for hysteroscopists.

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

2021 ◽  
Author(s):  
Bo Wang ◽  
Xiaoting Yu ◽  
Chengeng Huang ◽  
Qinghong Sheng ◽  
Yuanyuan Wang ◽  
...  
Keyword(s):  
Neural Networks ◽  
Classification Performance ◽  
Image Features ◽  
Deep Convolutional Neural Networks ◽  
Edge Information ◽  
Average Accuracy ◽  
Ship Classification ◽  
Edge Features ◽  
High Level ◽  
Better Than

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.

scholarly journals Double attention recurrent convolution neural network for answer selection

2020 ◽  
Vol 7 (5) ◽  
pp. 191517
Author(s):  
Ganchao Bao ◽  
Yuan Wei ◽  
Xin Sun ◽  
Hongli Zhang
Keyword(s):  
Neural Network ◽  
Question Answering ◽  
Convolution Neural Network ◽  
Deep Model ◽  
Selection Tasks ◽  
Matching Score ◽  
Key Steps ◽  
Chinese And English ◽  
Design Inspiration ◽  
Better Than

Answer selection is one of the key steps in many question answering (QA) applications. In this paper, a new deep model with two kinds of attention is proposed for answer selection: the double attention recurrent convolution neural network (DARCNN). Double attention means self-attention and cross-attention. The design inspiration of this model came from the transformer in the domain of machine translation. Self-attention can directly calculate dependencies between words regardless of the distance. However, self-attention ignores the distinction between its surrounding words and other words. Thus, we design a decay self-attention that prioritizes local words in a sentence. In addition, cross-attention is established to achieve interaction between question and candidate answer. With the outputs of self-attention and decay self-attention, we can get two kinds of interactive information via cross-attention. Finally, using the feature vectors of the question and answer, elementwise multiplication is used to combine with them and multilayer perceptron is used to predict the matching score. Experimental results on four QA datasets containing Chinese and English show that DARCNN performs better than other answer selection models, thereby demonstrating the effectiveness of self-attention, decay self-attention and cross-attention in answer selection tasks.

Overview

2011 ◽  
pp. 1-23
Author(s):  
David Zhang ◽  
Fengxi Song ◽  
Yong Xu ◽  
Zhizhen Liang
Keyword(s):  
Pattern Recognition ◽  
Recognition Performance ◽  
Recognition System ◽  
Biometric System ◽  
Biometric Recognition ◽  
Biometric Data ◽  
Pattern Recognition System ◽  
Biometric Systems ◽  
History Of

A biometric system can be regarded as a pattern recognition system. In this chapter, we discuss two advanced pattern recognition technologies for biometric recognition, biometric data discrimination and multi-biometrics, to enhance the recognition performance of biometric systems. In Section 1.1, we discuss the necessity, importance, and applications of biometric recognition technology. A brief introduction of main biometric recognition technologies are presented in Section 1.2. In Section 1.3, we describe two advanced biometric recognition technologies, biometric data discrimination and multi-biometric technologies. Section 1.4 outlines the history of related work and highlights the content of each chapter of this book.

Sign in / Sign up

Export Citation Format

Share Document