Intelligent Recognition System Research Articles

Image-Based Fitness Yoga Pose Recognition: Using Ensemble Learning and Multi-head Attention

With the increasing awareness of fitness, more and more people are choosing to participate in fitness activities. Yoga, as a form of exercise that improves both physical and mental health, is becoming increasingly popular worldwide. In order to assist yoga practitioners in more effective training through automated or semi automated systems, improve training effectiveness, assist professional athletes in training through intelligent recognition systems, correct movements, and improve athletic performance. This paper proposes a method that addresses the low accuracy issue of current yoga pose recognition algorithms by integrating multi-head attention mechanism and ensemble learning. Firstly, the Mixup algorithm is used to enhance yoga movement images. Subsequently, convolutional features are extracted from the images using the ResNet101 and VGGNet19 transfer learning models. Finally, the extracted convolutional features are combined and stacked using a multi-head attention mechanism. Model training, validation, and testing are performed using the Soft target cross-entropy loss function. Experimental results demonstrate that the proposed method achieves a training accuracy of 100%, a validation accuracy of 89.94%, a testing accuracy of 93.79%, and a detection speed of 297 frames per second. Overall, this method demonstrates high stability and robustness, providing a technological foundation for intelligent recognition of yoga poses.

Multifunctional Nano‐Conductive Hydrogels With High Mechanical Strength, Toughness and Fatigue Resistance as Self‐Powered Wearable Sensors and Deep Learning‐Assisted Recognition System

AbstractHigh mechanical strength, toughness, and fatigue resistance are essential to improve the reliability of conductive hydrogels for self‐powered sensing. However, achieving mutually exclusive properties simultaneously remains challenging. Hence, a novel directed interlocking strategy based on topological network structure and mechanical training is proposed to construct tough hydrogels by optimizing the network structure and modulating the orientation of molecular chains. Combining Zn2+ crosslinked cellulose nanofibers (CNFs) and a polyacrylamide‐poly(vinyl alcohol) double‐network, the unique interlocked‐network structure exhibits an enhanced toughening effect due to hydrogen bonding and metal‐ligand interactions. The aligned nanocrystalline domains achieved by training further contribute to an increase in the toughness and fatigue thresholds. This innovative approach synergistically enhances the mechanical properties of the nano‐conductive hydrogel, achieving a maximum tensile strength of 4.98 MPa and a toughness of 48 MJ m−3. Notably, the CNFs template with anchored polyaniline, when oriented through mechanical training, forms a unique directional conductive pathway, which significantly enhances the power output performance. Besides, a motion recognition system based on a self‐powered sensing device is designed with the assistance of deep learning techniques to accurately identify human motion behaviors. This work showcases a potentially transformative flexible electronic material for self‐powered sensing systems and intelligent recognition systems.

A Comparative Study on the Application of Robotic Hair Restoration Technology Versus Traditional Follicular Unit Excision in Male Androgenetic Alopecia

ABSTRACTBackgroundThe robotic hair transplant technology, ARTAS, has a series of fine mechanical structure and an intelligent recognition system that allows it to independently select hair follicular units (FUs) and effectively harvest hair. After entering China in 2016, ARTAS has attracted the attention of hair transplant surgeons and hair loss patients given its advantages in a short learning curve and simple operation.ObjectiveTo compare the efficacy and safety between the ARTAS system and follicular unit excision (FUE) in the treatment of male androgenetic alopecia (AGA) and to further promote the improvement and upgrading of ARTAS technology concerning hair transplantation.MethodsThirteen Chinese male patients with Norwood–Hamilton II–IV AGA aged 25–35 years were enrolled in this study. The donor site of each patient was randomly divided into left and right regions, receiving ARTAS on one side and FUE on the other. Yield, transection, and discard rates of hair FUs from both sides were compared. Safety of the procedures in whole, as well as follow‐up results were investigated and evaluated.ResultsThe total yield rate on the ARTAS side was lower than on the FUE side (82.05% vs. 90.03%, p > 0.05); the total discard rate on the ARTAS side was higher than on the FUE side (10.71% vs. 5.46%, p < 0.05); the total transection rate on the ARTAS side was lower than on the FUE side (13.17% vs. 13.96%, p > 0.05). No significant difference was found in patient satisfaction (efficacy), and no side effects or complications were detected during or after all surgeries.ConclusionThe current iteration of Robotic hair transplant technology is effective and safe, and can be recommended for AGA hair transplantation surgery.

Live Dimension Detection During Machining of Workpiece using Smart Detection Set up on General-Purpose Lathe

In a manufacturing setting, the profile dimensions of the workpiece are key quality factors. Various automatic inspection approaches currently assess dimensions based on characteristics and variables, tool paths as OK or Not OK in CNC, and online inspection methods for other machines after manufacturing. However, continuous on-machine profile dimensions and simultaneous manufacturing were left to focus. An intelligent recognition system based on a laser lines sensor using a smart profile dimension detection algorithm (SPDA) communicates real-time measurement information to the operator. It plots the current status of the Workpiece, mapping linear distance by accurate laser reflection of the exact mid-surface line of the workpiece as profile data of a symmetric shaft. It compares the length-diameter obtained from longitudinal and lateral sensors with the master data provided. The algorithm implements Python libraries for displaying panels and organizing workpieces. The standard image changes color from red to green as it is relevant to the status. The predetermined order of manufacturing parts and gradient descent optimization feature sets benchmarks to display the status of the workpiece. The real-time live measurement is facilitated with an error of less than 10%, i.e., 0.1 mm, saving manufacturing time by 30%. It also eliminates the inspection stage, and no reduced job rejection is found.

Developing a technology for modeling radar portraits of complex-shape objects for intelligent recognition systems

The object of this study is the modeling of radar portraits (RPs) for intelligent recognition systems based on the use of faceted 3D models. In order to solve the problems of target identification in homing systems of high-precision missile weapons, a technology is needed that could make it possible to efficiently and quickly generate RPs of military objects of complex shape in the required quantity. The research results are based on a combination of separate component technologies, in particular: the devised technology of using faceted 3D models – their construction and further processing with invisible surfaces excluded from it for an arbitrary viewing angle. The basic part of the work is the development of an algorithm and technological procedures for the formation of a spatial tracing grid for the current observation angle. A feature of the proposed technology is the application of a facet selection algorithm using an array of tracing facets and the application of the Huygens-Fresnel principle to recognize objects of complex shape. The RP database of military objects of complex shape was built. The results of modeling faceted RP’s, in particular the armored boat "Gyurza-M", are given. The results of the experimental study showed the ability to recognize the type of military object of complex shape at the level of 80‒90 %, which makes the use of this technology appropriate for recognizing military objects of complex shape. The achieved high-speed and quality characteristics of RP generation of military objects of a complex shape makes it possible to assume that the main prospective field of practical application is the identification and visual interpretation of targets in homing systems of high-precision missile weapons

A Recurrent Neural Network on Video-Based Face Identification Suspicious and Also Wanted People Recognition in ATM

The need for an intelligent recognition system is still increasing. Traditional approaches based on access to restricted places like ATM or suspected actions as theft, scam, and loitering. They are insufficient to identify suspects. These actions do not represent a real key of suspects. This project is motivated not only by the limits of the traditional approaches but also by the complexity of Intelligent Algorithms (RNN). In this project, we propose an approach for the automatic comparative labeling of facial soft biometric. The three main categories wanted persons, regular customers, and new persons serve distinct purposes. For wanted people recognition, the system is trained to identify individuals with outstanding or flagged in law enforcement databases. Regular ATM customers are recognized based on their registered facial features. New persons, on the other hand, are those not previously registered, and the system is designed to capture their facial data for future reference. Using a subset from the RNN-network datasets, recurrent neural network our experiments show the efficacy of the automatic generation of comparative facial labels, highlighting the potential extensibility of the approach to other face recognition scenarios and larger ranges of attributes. Recurrent Neural Network (RNN) system can be implement on CCTV cameras and it will be alerted particular officers. The application is easy to use and it is equipped with pyttsx3 so that the face detected is communicated to the people as voice alert.

Research on Intelligent Recognition and Verification System of Football Offside Penalty Based on Artificial Intelligence

An automatic recognition method for ball trajectory and penalty is extensively studied, leveraging computer vision technology. The core hardware components of this system are designed to effectively filter out noise and extract crucial information from the moving track. This enables the system to accurately identify the speed and position of the ball in real-time. To enhance the precision of image segmentation, a novel approach is introduced that utilizes the threshold vector in high-resolution color spaces. This method determines the color of each pixel through bitwise "and" operations between pixel points and subsets. Additionally, a moving window image filtering algorithm is proposed, incorporating trajectory prediction theory. Experimental results demonstrate that this approach significantly reduces recognition errors, resulting in a notably higher accuracy level, making it a promising solution for enhancing the performance of soccer robot systems.

Design and Implementation of Apple Leaf Disease Recognition System Based on ResNet50

Timely identification of apple leaf diseases is critical to preventing crop losses and safeguarding yields. Spotted leaf drop, brown spot, gray spot, mosaic and rust are all common types of apple leaf diseases, and their presence signals a potential risk that could lead to significant reductions in fruit and crop yields. The apple industry is thus at risk of economic losses. In order to solve this problem, this study applies ResNet50, a deep learning model for image recognition and classification of apple leaf diseases, and develops an intelligent recognition system for apple leaf diseases by combining PyQt technology. The purpose of this system is to overcome the shortcomings of traditional recognition methods. Through experimental validation, the ResNet50 model achieves a high accuracy rate of 93.19% in apple leaf disease recognition, demonstrating its efficiency and practicality in practical applications.

A facile fabrication strategy constructed multilayer piezoresistive pressure sensor for intelligent recognition system towards privacy protection

Privacy and security concerns have become a focal point of attention for humans, so that it is crucial to develop a secure and efficient identity recognition smart system. This work proposes a piezoresistive pressure sensor (PPS) based intelligent recognition system for privacy protection, wherein the PPS consists of multiple layers of electrospun fiber membranes with a bionic pine needle shaped structure. The PPS exhibits high sensitivity of 9040.46 kPa−1 at 0–1 kPa, 679.7 kPa−1 at 1–7 kPa, and 50.04 kPa−1 at 7–20 kPa, with a fast response/recovery time of 30 ms and excellent stability. Given these prominent features, the specific application in privacy protection has been construced. PPS was firstly exploited as an intelligent handwriting pad to obtain handwriting signals from different users and combined with convolutional neural network (LeNet-5) and long short-term memory (LSTM) network models to establish a handwriting recognition intelligent system for accurate identification of user identities. Further, a password unlocking control system was succesfully created based on the PPS array and support vector machine (SVM) network model for dual verification of passwords and behavior patterns. This work can easily and effectively detect people’s daily behavioral characteristics for identity authentication, and has enormous application potential in fields such as handwriting recognition, security defense, and privacy protection.

Adversarial Patch Attack Method for Vehicle Target Detection

Due to the susceptibility of intelligent recognition systems to adversarial sample attacks, research on adversarial sample attack methods for target detection is of great significance for improving the safety of vehicle autonomous driving. By studying the transfer-based self-ensemble attack adversarial sample generation algorithm, and based on the trained Faster R-CNN and YOLOX white box target detectors, a cross-model and cross-instance vehicle target detection attack method was designed. Through training, a universal adversarial patch was obtained, which achieved recognition attacks on two types of target detectors and three types of vehicles, resulting in incorrect recognition results being output. Through testing, the average decrease in mAP was about 41.6%, The research results have certain reference significance for the study of autonomous driving safety.

Design and Application Research on Modular Construction Waste Sorting, Recycling Equipment and Intelligent Recognition System

With the rapid advancement of urbanization and the continuous development of the construction industry, the generation of construction waste is increasing steadily, posing severe challenges to environmental protection and resource recycling. To address this issue, this paper delves deeply into the design and application of modular equipment for sorting and recycling construction waste, as well as intelligent recognition systems. This equipment system integrates advanced robotic arm technology, conveyor belt systems, electromagnetic suction rods, and cutting-edge visual recognition technology, with the aim of achieving intelligent sorting and efficient recycling of construction waste. This paper provides a detailed elaboration from multiple perspectives, including the system’s design philosophy, key technology applications, analysis of experimental results, and future development prospects. The objective is to offer innovative solutions for the resource utilization of construction waste.

A Robust Intelligent System for Text-Based Traffic Signs Detection and Recognition in Challenging Weather Conditions

A Robust Intelligent System for Text-Based Traffic Signs Detection and Recognition in Challenging Weather Conditions

Optimized Design of Instrument Recognition Based on CNN Model

Abstract Intelligent recognition of instrument features plays an important role in automation management and overhaul and also facilitates the realization of accurate reading of key parameters in complex environments. The instrument dial intelligent recognition system proposed in this paper consists of geometry correction, pointer segmentation, and reading recognition modules. Combining the idea of the GhostNet model to improve the structure of the backbone network of the Mask RCNN model, the attention mechanism is introduced into the U-Net model, and the minimum outer rectangle method is used for reading recognition. Under different viewpoint rotation angles, the recognition errors of this paper’s method are relatively stable, and they are less than 1%. The region segmentation precision, recall, and accuracy are 99.39%, 99.05%, and 98.38%, respectively. The average error of the recognition results is only -0.04°C, which is satisfactory for instrument recognition.

Kirigami‐Inspired 3D‐Printable MXene Organohydrogels for Soft Electronics

AbstractConductive hydrogels are compelling materials for the development of soft electronics; however, their essential attributes such as high sensitivity, excellent stretchability, and environmental stability have rarely been achieved simultaneously in one hydrogel. Herein, a Kirigami‐inspired strategy is proposed to improve organohydrogel sensitivity without sacrificing their mechanical stretchability and environmental stability . The organohydrogels with multiple interpenetrating networks are synthesized by introducing sodium alginate nanofibrils and conductive MXene nanoflakes into polymer double networks infiltrated with glycerol–water mixtures, featuring remarkable stretchability (>5000%), good sensitivity, and water retention (>30 days). The Kirigami structures are further applied to enhance strain sensitivity, achieving a gauge factor of 29.1, which is ≈5.5 times that of an unstructured organohydrogel. Using the Kirigami‐inspired sensors, a durable glove is developed for grabbing underwater objects through operating a robotic arm, demonstrating a subaqueous interactive human–machine interfacing.Meanwhile, by integrating the wearable sensor with a machine learning algorithm, a wearable Morse code intelligent recognition system is demonstrated, enabling real‐time conversion of Morse code signs into speech with superior recognition accuracy (>99%) and fast response time (≈17 ms). This work offers a new route to synthesize highly sensitive, stretchable, and extremely tolerant organohydrogels, providing a promising platform for next‐generation soft electronics.

An Intelligent System for Fingerprint Recognition and Verification Using Dilated Convolutional Neural Network and Extreme Learning Machine

Several applications currently utilize fingerprint-based user recognition and authentication, but obtaining complete accuracy (eliminating false matches) is still challenging. Prior to feature extraction, improper image alignment was one of the causes of this problem. This paper proposes a novel fingerprint recognition and verification system using a dilated convolutional neural network (DICNN) and a weighted and bias-optimized extreme learning machine (WBELM). Image preprocessing, feature extraction, and fingerprint recognition are the main phases of the proposed work. To begin, image preprocessing is performed using bilateral filtering to suppress the noise in the gathered images. After that, the proposed system uses DICNN to extract features from the preprocessed images. From the extracted features, fingerprint recognition is done using WBELM. This study conducted experiments on the publicly available FVC 2004 dataset. The proposed model achieves 98.54 percent recognition accuracy with an elapsed time of 584 ms.

Detection of various lung diseases including COVID-19 using extreme learning machine algorithm based on the features extracted from a lightweight CNN architecture

Around the world, several lung diseases such as pneumonia, cardiomegaly, and tuberculosis (TB) contribute to severe illness, hospitalization or even death, particularly for elderly and medically vulnerable patients. In the last few decades, several new types of lung-related diseases have taken the lives of millions of people, and COVID-19 has taken almost 6.27 million lives. To fight against lung diseases, timely and correct diagnosis with appropriate treatment is crucial in the current COVID-19 pandemic. In this study, an intelligent recognition system for seven lung diseases has been proposed based on machine learning (ML) techniques to aid the medical experts. Chest X-ray (CXR) images of lung diseases were collected from several publicly available databases. A lightweight convolutional neural network (CNN) has been used to extract characteristic features from the raw pixel values of the CXR images. The best feature subset has been identified using the Pearson Correlation Coefficient (PCC). Finally, the extreme learning machine (ELM) has been used to perform the classification task to assist faster learning and reduced computational complexity. The proposed CNN-PCC-ELM model achieved an accuracy of 96.22% with an Area Under Curve (AUC) of 99.48% for eight class classification. The outcomes from the proposed model demonstrated better performance than the existing state-of-the-art (SOTA) models in the case of COVID-19, pneumonia, and tuberculosis detection in both binary and multiclass classifications. For eight class classification, the proposed model achieved precision, recall and fi-score and ROC are 100%, 99%, 100% and 99.99% respectively for COVID-19 detection demonstrating its robustness. Therefore, the proposed model has overshadowed the existing pioneering models to accurately differentiate COVID-19 from the other lung diseases that can assist the medical physicians in treating the patient effectively.

Design and Realization of Computer Image Intelligent Recognition System

Design and Realization of Computer Image Intelligent Recognition System

Deep learning‐based recognition method of red bed soft rock image

In order to improve the investigation efficiency, improve the traditional engineering investigation work mode, realize the intelligent investigation, and improve the economic and social benefits of enterprises, a red soft rock image intelligent analysis and recognition system is proposed based on deep learning methods. The intelligent recognition system includes two core algorithms: soft rock image decomposition and soft rock lithology/weathering degree recognition. The research shows that the identification model of weathering degree and the lithology identification model based on the convolutional neural network (CNN) algorithm have a good identification effect, and the identification probability of moderately weathered rock reaches 95.22% and the accuracy of lithology identification is 91.34%. With the increase in training data, the recognition effect will be further improved. The intelligent identification system has been integrated into the WeChat mini programme, App, and Web system, which can be directly applied to field geological survey operations to assist geological workers in the investigation work, and realize the intelligent identification and classification of red bed soft rock.

Artificial Neural Network-Based Seedling Phenotypic Information Acquisition of Plant Factory

This work aims to construct an artificial neural network (ANN) ant colony algorithm (ACA)-based fine recognition system for plant factory seedling phenotypes. To address the problems of complexity and high delay of the plant recognition system in plant factories, first, multiple cameras at different positions are employed to collect images of seedlings and construct 3D images. Then, the mask region convolutional neural networks (MRCNN) algorithm is adopted to analyze plant phenotypes. Finally, the optimized ACA is employed to optimize the process timing in the plant factory, thereby constructing a plant factory seedling phenotype fine identification system via ANN combined with ACA. Moreover, the model performance is analyzed. The results show that plants have four stages of phenotypes, namely, the germination stage, seedling stage, rosette stage, and heading stage. The accuracy of the germination stage reaches 97.01%, and the required test time is 5.64 s. Additionally, the optimization accuracy of the process timing sequence of the proposed model algorithm is maintained at 90.26%, and the delay and energy consumption are stabilized at 20.17 ms and 17.71, respectively, when the data volume is 6000 Mb. However, the problem of image acquisition occlusion in the process of 3D image construction still needs further study. Therefore, the constructed ANN-ACA-based fine recognition system for plant seedling phenotypes can optimize the process timing in a more real-time and lower energy consumption way and provide a reference for the integrated progression of unmanned intelligent recognition systems and complete sets of equipment for plant plants in the later stage.

Polarization-sensitive narrowband infrared photodetection triggered by optical Tamm state engineering.

Polarization-sensitive narrowband photodetection at near-infrared (NIR) has attracted significant interest in optical communication, environmental monitoring, and intelligent recognition system. However, the current narrowband spectroscopy heavily relies on the extra filter or bulk spectrometer, which deviates from the miniaturization of on-chip integration. Recently, topological phenomena, such as the optical Tamm state (OTS), provided a new solution for developing functional photodetection, and we experimentally realized the device based on 2D material (graphene) for the first time to the best of our knowledge. Here, we demonstrate polarization-sensitive narrowband infrared photodetection in OTS coupled graphene devices, which are designed with the aid of the finite-difference time-domain (FDTD) method. The devices show narrowband response at NIR wavelengths empowered by the tunable Tamm state. The full width at half maximum (FWHM) of the response peak reaches ∼100 nm, and it can potentially be improved to ultra-narrow of about 10 nm by increasing the periods of dielectric distributed Bragg reflector (DBR). The responsivity and response time of the device reaches 187 mA/W and ∼290 µs at 1550 nm, respectively. Furthermore, the prominent anisotropic features and high dichroic ratios of ∼4.6 at 1300 nm and ∼2.5 at 1500 nm are achieved by integrating gold metasurfaces.