A human ear recognition method using nonlinear curvelet feature subspace

The use of the ear as a biometric modality has emerged in recent years. It makes it possible to differentiate people thanks to its stability over time and to the richness of its characteristics such as texture, color and size. This paper proposes a novel approach to ear recognition based on a variant of the Local Binary Pattern descriptor called Multi-scale Local Binary Pattern (MLBP). MLBP is calculated locally, by dividing the image into several equal blocks, to extract the ear features which will be used in the matching process to make a decision by detecting the similarities between the feature vectors using City-Block distance (CTB). The proposed method is evaluated on three reference ear databases: IIT Delhi I, IIT Delhi II and USTB-1. The analysis of the results obtained have clearly shown the robustness and the stability of the proposed ear recognition method which is highly competitive, achieving an attractive recognition performances in terms of identification rate at rank-1 up to: 98.40% for IIT Delhi I, 98.64% for IIT Delhi II, and 98.33% for USTB-1.

Three-dimensional data are increasingly being used for biometric purposes as they offer resilience to problems common mon in two-dimensional data. They have been successfully applied to face recognition and more recently to ear recognition. However, real-life biometric applications require algorithms that are both robust and efficient so that they scale well with the size of the databases. A novel ear recognition method is presented that uses a generic annotated ear model to register and fit each ear dataset. Then a compact biometric signature is extracted that retains 3D information. The proposed method is evaluated using the largest publicly available 3D ear database appended with our own database, resulting in a database containing data from multiple 3D sensor types. Using this database it is shown that the proposed method is not only robust, accurate and sensor invariant but also extremely efficient, thus making it suitable for real-life biometric applications.

A new optimal scheme based on curvelet transform is proposed for retinal image enhancement (RIE) using real-coded quantum genetic algorithm. Curvelet transform has better performance in representing edges than classical wavelet transform for its anisotropy and directional decomposition capabilities. For more precise reconstruction and better visualization, curvelet coefficients in corresponding subbands are modified by using a nonlinear enhancement mapping function. An automatic method is presented for selecting optimal parameter settings of the nonlinear mapping function via quantum genetic search strategy. The performance measures used in this paper provide some quantitative comparison among different RIE methods. The proposed method is tested on the DRIVE and STARE retinal databases and compared with some popular image enhancement methods. The experimental results demonstrate that proposed method can provide superior enhanced retinal image in terms of several image quantitative evaluation indexes.

Ear recognition is a complex research domain within biometrics, aiming to identify individuals using their ears in uncontrolled conditions. Despite the exceptional performance of convolutional neural networks (CNNs) in various applications, the efficacy of deep ear recognition systems is nascent. This paper proposes a two-step ear recognition approach. The initial step employs deep convolutional generative adversarial networks (DCGANs) to enhance ear images. This involves the colorization of grayscale images and the enhancement of dark shades, addressing visual imperfections. Subsequently, a feature extraction and classification technique, referred to as Mean-CAM-CNN, is introduced. This technique leverages mean-class activation maps in conjunction with CNNs. The Mean-CAM approach directs the CNN to focus specifically on relevant information, extracting and assessing only significant regions within the entire image. The process involves the implementation of a mask to selectively crop the pertinent area of the image. The cropped region is then utilized to train a CNN for discriminative classification. Extensive evaluations were conducted using two ear recognition datasets: mathematical analysis of images (MAI) and annotated web ears (AWEs). The experimental results indicate that the proposed approach shows notable improvements and competitive performance: the Rank-1 recognition rates are 100.00% and 76.25% for MAI and AWE datasets, respectively.

Video steganography mainly deals with secret data transmission in a carrier video file without being visually noticeable by intruders. Video steganography is preferred over image steganography because a video carries more space in comparison to an image. The main concept of information hiding consists of a cover media, which is a greyscale or a color video, a secret data, which is an image or text, and a stego key. Here a secure video steganography method has been proposed which uses Curvelet Transform for secret data embedding, Elliptic Curve Cryptography for stego key encryption and a threshold algorithm for the determination of the amount of secret data to be encoded per frame. A video is a collection of various frames. The frames are selected randomly from the cover video and the frame number of the respective frames has been indexed in the stego key to find the secret data embedding location. Here, the selection of frames in a sequential manner has been avoided to improve security. For enhanced security, the stego key is also encrypted using Elliptic Curve Integrated Encryption Scheme (ECIES). Fast Discrete Curvelet Transform (FDCT) has been applied to the frames of the cover video and the curvelet coefficients have been modified to obscure the secret data to produce the stego video.

This chapter presents an efficient ear recognition technique which derives benefits from the local features of the ear and attempt to handle the problems due to pose, poor contrast, change in illumination and lack of registration. It uses (1) three image enhancement techniques in parallel to neutralize the effect of poor contrast, noise and illumination, (2) a local feature extraction technique (SURF) on enhanced images to minimize the effect of pose variations and poor image registration. SURF feature extraction is carried out on enhanced images to obtain three sets of local features, one for each enhanced image. Three nearest neighbor classifiers are trained on these three sets of features. Matching scores generated by all three classifiers are fused for final decision. The technique has been evaluated on two public databases, namely IIT Kanpur ear database and University of Notre Dame ear database (Collections E). Experimental results confirm that the use of proposed fusion significantly improves the recognition accuracy.KeywordsBiometricsEar RecognitionImage EnhancementFusion

Extraction and description of image features is an active research topic and important for several applications of computer vision field. This paper presents a new noise-tolerant and rotation-invariant local feature descriptor called robust local oriented patterns (RLOP). The proposed descriptor extracts local image structures utilizing edge directional information to provide rotation-invariant patterns and to be effective in noise and changing illumination situations. This is achieved by a non-linear amalgamation of two specific strategies; binarizing neighborhood pixels of a patch and assigning binomial weights in the same formula. In the encoding methodology, the neighboring pixels is binarized with respect to the mean value of pixels in an image patch of size 3 × 3 instead of the center pixel. Thus, the obtained codes are rotation-invariant and more robust against noise and other non-monotonic grayscale variations. Ear recognition is considered as the representative problem, where the ear involves localized patterns and textures. The proposed descriptor encodes all images’ pixels and the resulting RLOP-encoded image is divided into several regions. Histograms of the regions are constructed to estimate the distribution of features. Then, all histograms are concatenated together to form the final descriptor. The robustness and effectiveness of the proposed descriptor are evaluated through conducting several identification and verification experiments on four different ear databases: IIT Delhi-I, IIT Delhi-II, AMI, and AWE. It is observed that the proposed descriptor outperforms the state-of-the-art texture based approaches achieving a recognition rate of 98% on the average providing the best performance among the tested descriptors.

This paper presents an automatic enhancement method for SAR images based on the mirror-extended curvelet transform and genetic algorithm. Firstly, an improved gain function which integrates the speckle reduction with the feature enhancement is proposed to nonlinearly shrink and stretch the curvelet coefficients, and then the genetic algorithm (GA) is used to automatically adjust the parameters of the gain function. We propose an objective criterion for enhancement, and attempt to find the (near) optimal image according to the respective criterion. We employ the GA as a global search strategy for the best enhancement which has a satisfactory compromise between sharpening and smoothing. The experimental results show that the proposed method can efficiently enhance the edge features and contrast of SAR images and reduce the speckle noises, and outperforms the wavelet- and curvelet-based non-automatic enhancement methods.

Ear recognition is an emerging research area in image-based biometrics. The commercial availability of lenslet light field cameras able to capture full spatio-angular information has brought momentum to biometric and forensic research exploiting this new type of imaging sensors. This study is the first to consider the usage of light field cameras for ear recognition, proposing both an appropriate content database and an ear recognition solution. The novel Lenslet Light Field Ear DataBase (LLFEDB) includes 536 light field images corresponding to four different poses, from 67 subjects, captured with a Lytro ILLUM lenslet light field camera. The LLFEDB includes critical cases such as ear images partly occluded by ear piercing, earing, hair and combination of occlusions. The novel light field-based ear recognition solution proposed exploits the richer spatio-angular information available in the light field images. A comparative performance evaluation study using the LLFEDB, and focusing on the most recent light field and non-light field based descriptors for ear recognition, shows a very promising performance of the proposed descriptor, outperforming all the assessed descriptors.

A novel method of feature-level fusion based on kernel Fisher discriminant analysis (KFDA) is proposed and applied to fusion of ear and profile face biometrics in this paper. Ear recognition is proved to be a new and promising authentication technique. Because of ear's special physiological structure and location, it is reasonable to combine ear with profile face for recognition in such scenarios as frontal face images are not available. First, only the face profile-view images are captured for recognition. Then based on KFDA, three feature fusion rules are presented. With the rules, the fusion discriminant vectors of ear and profile face are established and nonlinear feature fusion projection could be implemented. The experimental results show that the method is efficient for feature-level fusion, and the multimodal recognition based on ear and profile face performs better than ear or profile face unimodal biometric recognition. The work provides a new effective approach of non-intrusive biometric recognition.

Mammography is the most customarily used prior procedures to diagnose breast cancer. The preprocessing step is essential to improve the contrast in the mammogram image, in this paper top-hat and bottom-hat transforms are employed. This paper proposes the feature extraction method using curvelet transform in digital mammogram to distinguish the normal and abnormal breast cancer. The curvelet handles curves by utilising only a small quantity of coefficients, and it can also perform the curve discontinuities. It is a multiscale directional transform that permits sparse representation of objects with edges. The pectoral muscles and labels have been removed by the selection of Region of Interest (ROI). The proposed work uses morphological transform for preprocessing, followed by application of curvelet transform to enhance and sharpen the images finally wrapping is done by using fast Fourier transform. The features were extracted from the curvelet coefficients by applying gray level co-occurrance matrix and features were also extracted from the regional properties of the preprocessed image. The experimental result gives 98% accuracy of the classification of cancer classes using SVM classifier. From the result, it indicates that a curvelet transformation is a capable tool for analysis of digital mammogram.

An ear biometric system based on artificial bees and the scale invariant feature transform

Extraction and expression of features are critical to improving the recognition rate of ear image recognition. This paper proposes a new ear recognition method based on SIFT (Scale-invariant feature transform) and Forstner corner detection technology. Firstly, Forstner corner points and SIFT keypoints are detected respectively. Then taking Forstner corner into the SIFT algorithm to calculate their descriptor as the image feature vectors. Finally ear recognition based on these feature is carried out with Euclidean distance as similarity measurement. A bi-directional matching algorithm is utilized for improving recognition rate. Experiments on USTB database show that the recognition rate reaches more 94%. The Experimental results prove the effectiveness of the proposed method in term of recognition accuracy in comparison with previous methods. It is robust to rigid changes of ear image and provides a new approach to the research for ear recognition. DOI: http://dx.doi.org/10.11591/telkomnika.v11i12.2760

Recently biometric systems have shown improved capabilities because of the remarkable success of deep learning in solving various computer vision tasks. In ear recognition, the use of deep learning techniques is seldom due to training data scarcity. The existing work has shown poor performance as the majority of techniques are based on either handcraft features or pre-trained models. Besides this, transfer-learning has also shown poor performance because of the diversity among the tasks. To circumvent the existing issues, in this work, we have presented an end-to-end framework for ear recognition. It consist of the Ear Mask Extraction (EME) network to segment the ear, a normalization algorithm to align the ear, and a novel siamese-based CNN (CG-ERNet) for deep ear feature learning. CG-ERNet exploits domain-specific knowledge by using Curvature Gabor filters and uses triplet loss, triplet selection, and adaptive margin for better convergence of the loss. For comparative analysis, we trained state-of-the-art deep learning models like Face-Net, VGG19, ResNet50, Inception, Exception, and Mobile-Net for ear-recognition. The performance is assessed using five well-known evaluation metrics. In the extensive experimentation, our proposed model (CG-ERNet) outperformed the deep learning models and handcrafted feature based methods on four different, publicly available, benchmark datasets. To make the results more interpretable, we employ the t-SNE visualization of learned features. Additionally, our proposed method has shown robustness to various environmental challenges like Gaussian noise, Gaussian blur, up to ± 30 degrees of rotation, and 20% of occlusion.

Symmetrical null space LDA for face and ear recognition