Figure - available from: The Visual Computer
This content is subject to copyright. Terms and conditions apply.
Human Ear Anatomy. The human ear has very distinctive structural components. The outer ear is dominated by the shape of helix rim, lobe. The inner ear has many prominent features like antihelix, incisura intertragica, concha, triangular fossa, crus of helix, and tragus
Source publication
Human recognition systems based on biometrics are much in demand due to increasing concerns of security and privacy. The human ear is unique and useful for recognition. It offers numerous advantages over popular biometrics traits face, iris, and fingerprints. A lot of work has been attributed to ear biometric, and the existing methods have achieved...
Similar publications
We conducted more than 1.3 million comparisons of iris patterns encoded from images collected at two Nigerian universities, which constitute the newly available African Human Iris (AFHIRIS) database. The purpose was to discover whether ethnic differences in iris structure and appearance such as the textural feature size, as contrasted with an all-C...
Un glaucoma facogénico congénito es aquel desarrollado a raíz de la existencia de una patología subyacente en el cristalino al nacer, ya sea manifiesta en ese momento, o a lo largo de la vida. A diferencia del glaucoma congénito primario, aquí no existe una disgenesia del ángulo (la morfología del segmento anterior es normal), sino una predisposici...
Paper aims: The objectives of this paper are: (i) to analyze the impacts of Inventory Record Inaccuracy (IRI) on picking
productivity (PP), lost sales (LS), and warehouse capacity utilization (WCU) for different warehouses; (ii) verify if the Cycle
Counting (CC) implementation is sufficient to reduce IRI and (iii) the number of CC operators to ma...
![The basic module in measurement-based quantum computation [43, 44],...](publication/378153983/figure/fig1/AS:11431281223575233@1707796184615/The-basic-module-in-measurement-based-quantum-computation-43-44-where-Mth-denotes-the_Q320.jpg)
![The basic module in blind quantum computation [44], where Mθ1 denotes...](publication/378153983/figure/fig2/AS:11431281223517363@1707796184769/The-basic-module-in-blind-quantum-computation-44-where-Mth1-denotes-the-measurement_Q320.jpg)
It takes a lot of computational resources to train a machine learning model. So does quantum machine learning. In the NISQ (Noisy Intermediate Scale Quantum) era, there is a need for users who cannot afford quantum computers to utilize quantum servers to complete quantum machine learning with protection privacy of data and model parameters. In this...
Citations
... The ear as a biometric system has roots in 1890 with French criminologist Alphonse Bertillon, who proposed the first ear-based biometric system for identifying a person. He stressed that the ear is important for identification and recognition [3,4]. Subsequently, in the United States of America, in 1989, A. Iannarelli conducted further research by collecting 10,000 images of different ears and studied them using the manual measurement of twelve The ear as a biometric system has roots in 1890 with French criminologist Alphonse Bertillon, who proposed the first ear-based biometric system for identifying a person. ...
... Subsequently, in the United States of America, in 1989, A. Iannarelli conducted further research by collecting 10,000 images of different ears and studied them using the manual measurement of twelve The ear as a biometric system has roots in 1890 with French criminologist Alphonse Bertillon, who proposed the first ear-based biometric system for identifying a person. He stressed that the ear is important for identification and recognition [3,4]. Subsequently, in the United States of America, in 1989, A. Iannarelli conducted further research by collecting 10,000 images of different ears and studied them using the manual measurement of twelve distances as characteristics of the ear shape, which could uniquely identify people and concluded that the human ear is unique to each individual [2,5,6]. ...
... Figure 1. Images of biometric systems [3]. ...
One of the fundamental stages in recognizing people by their ears, which most works omit, is locating the area of interest. The sets of images used for experiments generally contain only the ear, which is not appropriate for application in a real environment, where the visual field may contain part of or the entire face, a human body, or objects other than the ear. Therefore, determining the exact area where the ear is located is complicated, mainly in uncontrolled environments. This paper proposes a method for ear localization in controlled and uncontrolled environments using MediaPipe, a tool for face localization, and YOLOv5s architecture for detecting the ear. The proposed method first determines whether there are cues that indicate that a face exists in an image, and then, using the MediaPipe facial mesh, the points where an ear potentially exists are obtained. The extracted points are employed to determine the ear length based on the proportions of the human body proposed by Leonardo Da Vinci. Once the dimensions of the ear are obtained, the delimitation of the area of interest is carried out. If the required elements are not found, the model uses the YOLOv5s architecture module, trained to recognize ears in controlled environments. We employed four datasets for testing (i) In-the-wild Ear Database, (ii) IIT Delhi Ear Database, (iii) AMI Ear Database, and (iv) EarVN1.0. Also, we used images from the Internet and some acquired using a Redmi Note 11 cell phone camera. An accuracy of 97% with an error of 3% was obtained with the proposed method, which is a competitive measure considering that tests were conducted in controlled and uncontrolled environments, unlike state-of-the-art methods.
... Nowadays, biometric features are mainly computed through the application of a deep network. We can cite face descriptors (Wen et al., 2016), ear (Kamboj et al., 2022) or palm print (Trabelsi et al., 2022). In this paper, we consider having a CNN that generates fixed size feature vectors as a representation of the biometric modality. ...
... There are several types of biometrics that can be encapsulated, evaluated, and measured to decide on identification. Some examples include DNA, signature, voice, speech, ear, palmprint, the shape of the hand (hand geometry), the layer of blood vessels at the back of the eye (retina), face, patterns of heat radiated by the human body (facial thermogram), the peculiar way an individual walks (gait), the coloured ring of tissue that surrounds the pupil (iris), the way a person types on the keyboard (keystroke), and the exuded chemical composition (odour) (Kamboj et al., 2021;Liu and Cao, 2012;Liu and Silverman, 2001). Table 1, inspired by the work of Jain et al. (2004) and Sabhanayagam et al. (2018), compares the various forms of biometrics from different perspectives such as universality, distinctiveness, collectability, performance, acceptability, and circumvention. ...
... The fingerprint being scanned is verified to be a live fingerprint by completing the handshake and applying the SET protocol commonly used in any secure communication. A similar approach is used in Kushwaha and Nain (2020) but for footprint and Kamboj et al. (2021) for ear-based recognition. The live fingerprint is then compared with the fingerprint stored in the smart card. ...
... Nevertheless, despite the potential of ear biometrics, there exist notable obstacles to implementing this technology in unregulated environments. The accuracy and reliability of ear-based recognition systems can be compromised by differences in characteristics such as position, scale, and occlusion, leading to the emergence of these difficulties (Kamboj et al. 2022). In order to tackle the aforementioned issues and effectively utilize the capabilities of ear biometrics, traditional DL and hybrid techniques are required (Karasulu et al. 2022). ...
Biometric-based personal authentication systems have experienced significant demand, primarily driven by growing concerns surrounding privacy and security in various applications. However, research has demonstrated that the human ear possesses sufficient distinguishing features to serve as a robust biometric measure. Ear biometrics present significant challenges in uncontrolled environments due to variations in factors such as pose, scale, and occlusion. To tackle the issue of human identification using ear images, the study proposes a framework named CSA-GRU based on a convolutional neural network (CNN) and gated recurrent unit (GRU) with self-attention. The current study employed four distinct datasets, namely IITD-I, IITD-II, AMI, and AWE, which underwent five augmentation techniques, including flipping, Gaussian noise, color jitter, brightness adjustment, and translation. Subsequently, a hybrid approach employing CNN and GRU with self-attention is used to extract features and perform human identification. The performance of CSA-GRU has been evaluated and yielded recognition rates of 99.24%, 99.81%, 99.07%, and 98.81% for the datasets mentioned above. Furthermore, a comparative examination has been conducted on CSA-GRU in correlation with diverse established methodologies. The findings indicate enhancements of 0.40%, 1.07%, 0.97%, and 18.30% across the above-mentioned datasets.
... Recognition systems have evolved as a grand challenge for machine learning, with the long-term goal of achieving near-human recognition levels for thousands of categories under a variety of aspects. Deep learning is a type of machine learning that has found use in a wide variety of research domains such as biometrics applications [15], as well as in other fields [16]- [18]. Transfer learning enables the usage of pre-trained networks by fine-tuning them with domain-specific data, hence increasing their effectiveness. ...
Agricultural images such as fruits and vegetables have previously been recognised and classified using image analysis and computer vision techniques. Mangoes are currently being classified manually, whereby mango sellers must laboriously identify mangoes by hand. This is time-consuming and tedious. In this work, TensorFlow Lite was used as a transfer learning tool. Transfer learning is a fast approach in resolving classification problems effectively using small datasets. This work involves six categories, where four mango types are classified (Harum Manis, Langra, Dasheri and Sindhri), categories for other types of mangoes, and a non-mango category. Each category dataset comprises 100 images, and is split 70/30 between the training and testing set, respectively. This work was undertaken with a mobile-based application that can be used to distinguish various types of mangoes based on the proposed transfer learning method. The results obtained from the conducted experiment show that adopted transfer learning can achieve an accuracy of 95% for mango recognition. A preliminary user acceptance survey was also carried out to investigate the user’s requirements, the effectiveness of the proposed functionalities, and the ease of use of its proposed interfaces, with promising results.
... Smart Home has been a feature of science fiction writings for many years but has only become practical since the early 20th century following the widespread introduction of electricity into the home, and the rapid advancement of information technology. Various projects have been developed for identity recognition by various biometric approaches [14]. However, the literature shows that the single-mode systems have limitations mainly due to noise acquisition, the problem of universality of certain biometrics, the possibilities of attacks, and the use of cultural limitations. ...
Since the smartphone is adapted to the ambient intelligence and the smart home systems are remotely accessed through the smartphones, there is a need for a secure authentication system based on some biometrics proprieties that can be taken from a smartphone. The identification of persons through ear and voice print is one of the basic biometric matters. The earlier research in ear recognition have shown that human ear is one of the representative human biometrics with uniqueness and stability. Indeed, the human voice is a perfect source of data for person identification in many applications. In this paper, we propose a fusion between the ear and voice biometrics in degraded conditions in a smart home context at 3 levels (feature, score, and decision). The experiments are conducted on the EVDDC database and a chimeric database (TIMIT and USTB-I). The best results are obtained with the feature level fusion (95.8%) with the KNN classifier.
... This filtering process allows only sounds relevant to humans, primarily for existential reasons, to pass through. As a result, terms like audible range, sound perception, pitch, rhythm, timbre, hearing threshold, correction curves, etc. [28][29][30][31][32][33][34] are commonly used in this context. The assessment of sound perception falls within the scientific field of psychoacoustics [35][36][37][38]. ...
The article aims to examine the influence of various acoustic materials on the usable properties of headphones. Acoustic materials act as filters that modify the incident acoustic waves by attenuating specific frequency components of the acoustic signals. They can therefore be called acoustic filters because they affect the frequency response (bandwidth) of the headphones in question. Therefore, the choice of acoustic materials in headphones with a specific design can have a significant impact on their psychoacoustic properties. Calibrating your headphones is critical to ensure they don't amplify, attenuate or introduce any coloration to your frequencies. To address this problem, a dedicated test setup was designed and constructed to evaluate the frequency response of selected headphone models using seven different acoustic materials (acoustic filters). The final selection of the acoustic filter was made using an artificial neural network. The article presents the measurement methodology, and the obtained frequency response measurements were analyzed and compared.
... A detailed review of the current DLbased articles for 3D pose estimation and a summary of the merit and demerit of those methods are presented [27]. Another study provides a review of current study on multi-person pose estimation and analyses the algorithms and compares their advantages and the disadvantages to fill in the gap of the existing surveys [28]. Most of the previous surveys focused on reviewing the DL approach in tackling HPE issues but did not consider summarizing and tabulating the development of DL in HPE from the year of breakthrough to date. ...
In this article, a comprehensive survey of deep learning-based (DL-based) human pose estimation (HPE) that can help researchers in the domainof computer vision is presented. HPE is among the fastest-growing researchdomains of computer vision and is used in solving several problems forhuman endeavours. After the detailed introduction, three different humanbody modes followed by the main stages of HPE and two pipelines of two-dimensional (2D) HPE are presented. The details of the four components ofHPE are also presented. The keypoints output format of two popular 2D HPEdatasets and the most cited DL-based HPE articles from the year of break-through are both shown in tabular form. This study intends to highlight thelimitations of published reviews and surveys respecting presenting a systematicreview of the current DL-based solution to the 2D HPE model. Furthermore,a detailed and meaningful survey that will guide new and existing researcherson DL-based 2D HPE models is achieved. Finally, some future researchdirections in the field of HPE, such as limited data on disabled persons andmulti-training DL-based models, are revealed to encourage researchers andpromote the growth of HPE research.
(PDF) A Survey on Deep Learning-Based 2D Human Pose Estimation Models. Available from: https://www.researchgate.net/publication/373611649_A_Survey_on_Deep_Learning-Based_2D_Human_Pose_Estimation_Models#fullTextFileContent [accessed Sep 05 2023].
... Gender identification has become a major concern due to its vast variety of applications, including social communication and connection, commercial visual supervision, banking transactions, illness prognosis, demographic data collection, artificial intelligence (AI) based user interface for customization, consumer analysis for business growth, and many more [17], [35]. Biometric traits have proven their suitability in gender classification as they are non-intrusive, remain invariant with time, and are less influenced by emotions, and circumstances. ...
Human gender classification based on biometric features is a major concern for computer vision due to its vast variety of applications. The human ear is popular among researchers as a soft biometric trait, because it is less affected by age or changing circumstances, and is non-intrusive. In this study, we have developed a deep convolutional neural network (CNN) model for automatic gender classification using the samples of ear images. The performance is evaluated using four cutting-edge pre-trained CNN models. In terms of trainable parameters, the proposed technique requires significantly less computational complexity. The proposed model has achieved 93% accuracy on the EarVN1.0 ear dataset.
... Despite recent considerable advancements, it is still difficult to create a reliable pedestrian detection solution that is ready for use in real-world applications. It has been noted that the majority of pedestrian detectors in use today are trained using just visible input, making their performances vulnerable to variations in lighting, weather, and occlusions [11][12][13][14][15][16]. Numerous studies have concentrated on creating multispectral pedestrian identification technologies to enable reliable human target detection for 24-h use in order to get over the aforementioned limitations. ...
Autonomous electric vehicle safety is crucially dependent on the accurate recognition of pedestrians in diverse situations. Current pedestrian detection techniques, however, face significant limitations due to reduced visibility and poor-quality images under low-lighting scenarios. With the aim of overcoming these challenges, this article proposes a novel, sustainable method for pedestrian detection and classification in electric vehicles using machine learning techniques. The approach processes video frame-based images as input, removing noise and smoothing the images for improved detection. A Bayesian component network analysis is employed to refine the features of the filtering-based boundary box detection, further enhancing the detection process. The selected features are then classified using a fully connected kernel operation based on the region with reward Q-Reinforcement architecture, resulting in a secure and efficient pedestrian detection system. The proposed method was evaluated on multiple image datasets using average precision, an area under the curve (AUC), log-average miss rate (MR), and root-mean-square error (RMSE) as performance measures. The experimental results demonstrated an average precision of 92%, MR of 48%, AUC of 56%, and RMSE of 61%. These findings indicate that the proposed technique effectively enhances pedestrian detection and classification for autonomous electric vehicles, contributing to increased safety and reliability in real-world applications.
