Sign Detection Research Articles

Sign Language Detection Using Machine Learning

This paper investigates the application of machine learning for sign language detection. The objective is to develop a model that translates sign language into spoken language, bridging the communication gap between deaf and hearing individuals. You Only Look Once (YOLO), a deep learning object detection algorithm, is employed to train a model on a dataset of labeled sign language images derived from video data. The system achieves real-time sign detection in videos. However, challenges include the scarcity of large, labeled datasets and the inherent ambiguity of certain signs, which can lead to reduced detection accuracy. This research contributes to the field of Assistive Technologies (AT) by promoting accessibility and social inclusion for the deaf community.

Traffic sign detection and recognition based on MMS data using YOLOv4-Tiny algorithm

Traffic sign detection and recognition based on MMS data using YOLOv4-Tiny algorithm

Research on traffic sign detection algorithm based on improved SSD in complex environments

Abstract In complex traffic sign environments, detection challenges include uneven size distribution, insufficient valid information, and difficulties in identifying targets under resource constraints, leading to missed and false detections. This study proposes an enhanced lightweight traffic sign detection algorithm based on SSD (Single Shot MultiBox Detector). By replacing the original backbone network with MobileNetV2, the model is streamlined to have fewer parameters, which improves generalization in complex environments. This modification significantly boosts the recall rate, achieving a better balance between precision and recall. By introducing the FPN(Feature Pyramid Network) combined with the CBAM(Convolutional Block Attention Module) attention mechanism, the detailed and semantic information between deep and shallow layers is fully integrated, reducing the loss of feature information, thus enhancing the strengthening of key information of traffic signs and the adaptability to different scales of traffic signs. Finally, by integrating the cross-attention mechanism, the algorithm's anti-interference ability in complex environments is improved, and the positioning accuracy of traffic signs is enhanced by capturing the dependency between different positions. Through ablation experiments and comparative experiments on a public traffic sign dataset, our improved SSD algorithm achieved an mAP of 89.97%. Compared with the original algorithm, the mAP increased by 12.41%, the recall rate increased by 18.38%, and the sum of precision and recall F1 increased by 14.6%. These improvements significantly enhance the performance of traffic sign detection in complex environments, thereby meeting the performance requirements of traffic sign detection.

Advancements in Traffic Sign Recognition and Detection: Harnessing the Power of Intelligent Systems

Advancements in Traffic Sign Recognition and Detection: Harnessing the Power of Intelligent Systems

A two-stage road sign detection and text recognition system based on YOLOv7

We developed a two-stage traffic sign recognition system to enhance safety and prevent tragic traffic incidents involving self-driving cars. In the first stage, YOLOv7 was employed as the detection model for identifying 31 types of traffic signs. Input images were set to 640 × 640 pixels to balance speed and accuracy, with high-definition images split into overlapping sub-images of the same size for training. The YOLOv7 model achieved a training accuracy of 99.2 % and demonstrated robustness across various scenes, earning a testing accuracy of 99 % in both YouTube and self-recorded driving videos. In the second stage, extracted road sign images underwent rectification before processing with OCR tools such as EasyOCR and PaddleOCR. Post-processing steps addressed potential confusion, particularly with city/town names. After extensive testing, the system achieved recognition rates of 97.5 % for alphabets and 99.4 % for Chinese characters. This system significantly enhances the ability of self-driving cars to detect and interpret traffic signs, thereby contributing to safer road travel.

Enhancing Road Safety: Fast and Accurate Noncontact Driver HRV Detection Based on Huber-Kalman and Autocorrelation Algorithms.

Enhancing road safety by monitoring a driver's physical condition is critical in both conventional and autonomous driving contexts. Our research focuses on a wireless intelligent sensor system that utilizes millimeter-wave (mmWave) radar to monitor heart rate variability (HRV) in drivers. By assessing HRV, the system can detect early signs of drowsiness and sudden medical emergencies, such as heart attacks, thereby preventing accidents. This is particularly vital for fully self-driving (FSD) systems, as it ensures control is not transferred to an impaired driver. The proposed system employs a 60 GHz frequency-modulated continuous wave (FMCW) radar placed behind the driver's seat. This article mainly describes how advanced signal processing methods, including the Huber-Kalman filtering algorithm, are applied to mitigate the impact of respiration on heart rate detection. Additionally, the autocorrelation algorithm enables fast detection of vital signs. Intensive experiments demonstrate the system's effectiveness in accurately monitoring HRV, highlighting its potential to enhance safety and reliability in both traditional and autonomous driving environments.

Real-time sign language detection: Empowering the disabled community

Interaction and communication for normal human beings are easier than for a person with disabilities like speaking and hearing who may face communication problems with other people. Sign Language helps reduce this communication gap between a normal and disabled person. The prior solutions proposed using several deep learning techniques, such as Convolutional Neural Networks, Support Vector Machines, and K-Nearest Neighbors, have either demonstrated low accuracy or have not been implemented as real-time working systems. This system addresses both issues effectively. This work extends the difficulties faced while classifying the characters in Indian Sign Language(ISL). It can identify a total of 23 hand poses of the ISL. The system uses a pre-trained VGG16 Convolution Neural Network(CNN) with an attention mechanism. The model's training uses the Adam optimizer and cross-entropy loss function. The results demonstrate the effectiveness of transfer learning for ISL classification, achieving an accuracy of 97.5 % with VGG16 and 99.8 % with VGG16 plus attention mechanism.•Enabling quick and accurate sign language recognition with the help of trained model VGG16 with an attention mechanism.•The system does not require any external gloves or sensors, which helps to eliminate the need for physical sensors while simplifying the process with reduced costs.•Real-time processing makes the system more helpful for people with speaking and hearing disabilities, making it easier for them to communicate with other humans.

Cutting-Edge Microwave Sensors for Vital Signs Detection and Precise Human Lung Water Level Measurement

In this article, a comprehensive review is presented of recent technological advancements utilizing electromagnetic sensors in the microwave range for detecting human vital signs and lung water levels. With the main objective of improving detection accuracy and system robustness, numerous advancements in front-end architecture, detection techniques, and system-level integration have been reported. The benefits of non-contact vital sign detection have garnered significant interest across a range of applications, including healthcare monitoring and search and rescue operations. Moreover, some integrated circuits and portable systems have lately been shown off. A comparative examination of various system architectures, baseband signal processing methods, system-level integration strategies, and possible applications are included in this article. Going forward, researchers will continue to focus on integrating radar chips to achieve compact form factors and employ advanced signal processing methods to further enhance detection accuracy.

Improved Arithmetic Optimization Algorithm with Transfer Learning based Arabic Sign Language Identification System

The Arabic sign language (ArSL) has witnessed ground-breaking research activities to identify hand gestures and signs through the deep learning (DL) model. SL is a unique communication tool that bridges the gaps between people with hearing impairment and ordinary people. The ArSL recognition system is of immense importance for different groups of people since it enables individuals with hearing impairment to communicate effectively. In SLs, signs are characterized by discrepancies in hand positions, shapes, motions, body parts, and facial expressions, posing a crucial threat to visual recognition in computer vision (CV). An automated sign detection technique needs two primary courses of action: the recognition of specific features and the classification of the input dataset. Previously, several approaches for detecting and classifying SLs had been proposed. In this study, an Improved Metaheuristics with Transfer Learning based Arabic Sign Language Identification System (IMTL-ArSL) method is developed. The primary objective of the IMTL-ArSL method is to detect and classify the existence of various signs in the Arabic language. In the IMTL-ArSL model, the bilateral filtering (BF) model is initially used for preprocessing. Besides, the Residual Network (ResNet50v2) model is applied for extracting features, and an improved arithmetic optimization algorithm (IAOA) model is utilized for hyperparameter tuning. Finally, a gated recurrent unit (GRU) network is exploited to identify sign languages. The empirical analysis of the IMTL-ArSL approach is tested using a benchmark dataset. The experimental values of the IMTL-ArSL approach portrayed a superior accuracy value of 93.87% over other techniques.

TLDM: An Enhanced Traffic Light Detection Model Based on YOLOv5

Traffic light detection and recognition are crucial for enhancing the security of unmanned systems. This study proposes a YOLOv5-based traffic light-detection algorithm to tackle the challenges posed by small targets and complex urban backgrounds. Initially, the Mosaic-9 method is employed to enhance the training dataset, thereby boosting the network’s ability to generalize and adapt to real-world scenarios. Furthermore, the Squeeze-and-Excitation (SE) attention mechanism is incorporated to improve the network. Moreover, the YOLOv5 algorithm’s loss function is optimized by substituting it with Efficient Intersection over Union loss (EIoU_loss), which addresses issues like missed detection and false alarms. Experimental results demonstrate that the model trained with this enhanced network achieves an mAP (mean average precision) of 99.4% on a custom dataset, which is 6.3% higher than that of the original YOLOv5, while maintaining a detection speed of 74 f/s. Therefore, this algorithm offers higher detection accuracy and effectively meets real-time operational requirements. The proposed method has a strong application potential, and can be widely used in the field of automatic driving, assisted driving, etc. Its application is not only of great significance in improving the accuracy and speed of traffic sign detection, but also can provide technical support for the development of intelligent transportation systems.

Níveis de CO2, quebra técnica e qualidade dos grãos de milho armazenados em diferentes condições

ABSTRACT The use of CO2 sensors has been reported as an effective tool in the early detection of signs of deterioration, allowing good accuracy in decision-making about the quality of stored grains. The objective of this study was to quantify the CO2 levels produced by maize grains stored at ambient temperature in a prototype silo, with initial moisture contents of 14, 16 and 18% w.b., and to evaluate the quality of the product over time and the technical breakage. Analyses of moisture content, ash, proteins, lipids, Hue Angle (color), germination and electrical conductivity were performed. Inside the silos, the amount of CO2, relative air humidity and temperature were monitored every hour. Grain quality evaluations were carried out at four storage times (0, 30, 60 and 90 days). A completely randomized design in a 3 × 4 factorial scheme with six replicates was used. Grains with higher moisture contents showed higher technical breakage and losses associated with CO2 emissions. Moisture content, protein, lipids, germination, bulk density, lightness and chroma decreased over time, while the electrical conductivity increased, resulting in greater damage to membranes and loss of quality of maize grains. Monitoring of CO2 in the grain mass is a good tool to assess the quality of grain, and higher initial moisture content caused greater level of CO2 emission and reduction in quality of maize grains.

Toward Effective Traffic Sign Detection via Two-Stage Fusion Neural Networks

Toward Effective Traffic Sign Detection via Two-Stage Fusion Neural Networks

AI Drive Assist: Enhancing Road Safety through Advanced Road Sign Detection and Driver Alerts

AI Drive Assist: Enhancing Road Safety through Advanced Road Sign Detection and Driver Alerts

DETECTION OF THE VITAL ORIGIN OF THE STRANGULATION MARK IN MECHANICAL ASPHYXIA AS AN INDICATOR OF THE DEVELOPMENT OF FORENSIC MEDICINE

Death by mechanical asphyxia occupies one of the leading positions in the structure of violent deaths. In the structure of mechanical asphyxia, one of the most common is mechanical strangulation, which includes hanging, ligature strangulation and manual strangulation. Solving the problem of the ante-mortem nature of mechanical asphyxia remains relevant in the practical work of forensic experts. In the case of mechanical asphyxia by strangulation, the most common and effective methods are the study of changes in the area of local injuries on the neck, such as the ligature mark during hanging. Aim of the study. To shed light on the process of developing methods for determining the ante-mortem nature of mechanical strangulation asphyxia by studying changes in the area of the ligature mark and to determine effective ways of solving this problem at the current stage of development of forensic medicine. Methods. Analysis and synthesis of information from literary sources available in online databases. Results. The use of macroscopic and histological methods to study changes in the area of the ligature mark may not be sufficient to resolve the issue of ante-mortem formation of injuries in the neck area during hanging, due to the extremely short period of time that elapses between the onset of the asphyxia process and the onset of death. The time interval between the onset of asphyxia and death is often insufficient for the development of a specific reactive cellular and even vascular response. Immunohistochemical examination, aimed at identifying initial reactive signs in the area of injury, is more effective in establishing ante-mortem status during extremely short time intervals between the cause of injury and the onset of death. In the case of mechanical asphyxia by strangulation, significant results have been obtained in the study of ligature marks by immunohistochemical markers such as tryptase, IL-15, CD-15, MHC-II, CD1a, AQP1, AQP3, fibronectin D, P-selectin. In addition, biochemical methods for the detection of physiologically active substances, which allow the detection of early signs of reactive changes in the skin due to injury, such as the study of the concentration of histamine, serotonin, heparin, prostaglandins F and E, as well as the ratio of certain ions, can be highly effective in diagnosing the ante-mortem nature of the ligature mark. Conclusion. The use of immunohistochemical research methods to confirm the ante-mortem formation of a ligature mark is based on the detection of markers of the inflammatory reaction, which are released before the appearance of any vascular and cellular reactions, allowing the detection of the first signs of the inflammatory reaction even within a very short period of time between the cause of injury and the onset of death. Thus, the use of immunohistochemical methods can be effective in confirming the ante-mortem nature of mechanical asphyxia, especially in cases where the use of standard histological staining methods has been ineffective. The detection of certain biochemical markers in the skin from the area of the ligature mark may be considered a promising, rapid and effective method to confirm the ante-mortem nature of strangulation asphyxia. This direction requires further research to improve its effectiveness and can be used in forensic practice to confirm the ante-mortem formation of the ligature mark.

Enhancing Sign Language Understanding through Machine Learning at the Sentence Level

Abstract: The visual language of sign language is based on nonverbal communication, including hand and body gestures. When it comes to communicating, it is the main tool for those who are deaf or hard of hearing all around the globe. Useful for both hearing and deaf persons, this can translate sign language into sentences in real-time or help those who are hard of hearing communicate with others. This work focuses on developing a sentence-level sign language detection system utilizing a custom dataset and Random Forest model. Leveraging tools such as Media Pipe and TensorFlow, we facilitate gesture detection. Through continuous detection of gestures, we generate a list of corresponding labels. These labels are then used to construct sentences automatically. The system seamlessly integrates with ChatGPT, allowing direct access to generate sentences based on the detected gestures. Our custom dataset ensures that the model can accurately interpret a wide range of sign language gestures. Our method helps close the communication gap between people who use sign language and others, with an accuracy of 80%, by merging machine learning with complex language models.

Prevention of vision disorders at school age: view of secondary school teachers on the problem

Background. The relevance of the study is due to the importance of maintaining full vision functions from childhood, which allows for having healthy visual perception for a quality life, learning and work capacity in adolescence and adulthood. The problem is gaining global importance, because according to the forecasts of international experts, the spread of ophthalmic pathology in 2050 will reach such a scale that almost a third of the planet’s population will have impaired vision, with most of cases diagnosed already in childhood. Scientific studies reveal an increase in the frequency of visual impairment among school-aged children. Thus, 8–10 % of the first-graders are diagnosed with ophthalmic pathology, and upon completion of schooling, this indicator can increase to 30–50 %. The purpose was to study the awareness of teachers and lecturers of secondary educational institutions (schools, lyceums) regarding the problems of visual impairment among schoolchildren, their attitude to the problem and willingness to participate in preventive measures. Materials and methods. The materials for the research were the results of a sociological survey of educators of secondary educational institutions from all regions of Ukraine (lecturers, teachers at schools and lyceums), a total of 6,215 questionnaires were subject to analy­sis. The tool for conducting sociological research was the deve­loped Questionnaire of secondary school educators regarding their awareness and willingness to participate in preventive programs to preserve eyesight from childhood. The procedure for filling out the questionnaire included an interview with an interviewer, who verbally informed about the purpose, tasks of sociological research, and its anonymity. Only after obtaining preliminary voluntary verbal consent to participate in the study, respondents were offered to proceed to filling out the questionnaire, which took an average of 15–20 minutes. The research methods were sociological survey, analytical-synthetic, medical-statistical methods, in particular the latter was used to calculate the representativeness and reliability of the survey results. Results. The results of the study prove the importance of the participation of secondary school educators in measures to prevent impairment of visual functions in school-aged children. Despite the high readiness of lecturers and secondary school teachers to engage in preventive activities, their awareness of preventing visual impairment among schoolchildren was found to be insufficient, as was the level of preventive work of medical personnel in schools. It was found that most educators are convinced that a school education worker should be aware of the signs of visual impairment in children (90.46 ± 0.37 %), participate in preventive measures and be familiar with the hygienic norms of visual stress in children (86.39 ± 0.43 %). Most respondents expressed the feasibility of introducing regular systematic supervision of vision functions in school-aged children (99.40 ± 0.10 %) and supporting the implementation of state preventive programs to preserve vision from childhood in Ukraine (97.38 ± 0.20 %), self-critically recognizing the need for informational support regarding the detection of signs and symptoms of these diseases in children (71.70 ± 0.57 %). Conclusions. Visual impairments in children of different ages are of great medical and social importance, as they affect the formation of disorders of socialization and human performance. The results of the study open the prospect of involving secondary school educators in the activities of interdisciplinary comprehensive programs to prevent the development of ophthalmic pathology among school-aged children. The search for effective ways of informational support for educators on the issues of modern preventive technologies, risk factors for the formation of visual impairments in schoolchildren in order to prevent the formation of stigma about children with disabilities and visual impairments in their environment requires scientific justification.

A Systematic Study Traffic Sign Detection System, Different Types of Road Traffic Sign and Various Techniques of Road Traffic Sign Recognition

A Systematic Study Traffic Sign Detection System, Different Types of Road Traffic Sign and Various Techniques of Road Traffic Sign Recognition

Lane Detection, Segmentation, Pothole Detection and Traffic Sign Recognition for ADAS

Lane detection, a critical aspect of advanced driver assistance systems (ADAS) and autonomous vehicles, is addressed in this work, combining classical computer vision methods with deep learning techniques. The proposed solution, utilizing a UNet-based model trained with TensorFlow and Keras, enhances vehicle perception for intelligent transportation systems. Integrated functionalities for pothole detection and traffic sign detection further contribute to safety and efficiency, enabling vehicles to identify road hazards and comply with regulations. The system encompasses data preprocessing, model training, and real-time video analysis, while a classical lane detection pipeline using OpenCV showcases various stages such as grayscale conversion, Gaussian blur, Canny edge detection, masking, Hough transform, and lane overlay. This comprehensive approach supports road safety, traffic management, and transportation efficiency initiatives, making significant strides in intelligent transportation systems and urban planning.

Classification of Traffic Signs Using Transfer Learning Methods

Transportation refers to a process based on the movement of people or vehicles from one place to another. Sea routes and roads have existed for centuries. They generally play a very important role in people's daily life, trade and industrial activities. Highway, a mode of transportation, is the first preferred mode of transportation worldwide. However, various signs and rules have been set by the authorities to prevent chaos on the highways. Traffic signs are the most important of these rules. In this study, transfer learning models (VGG16, VGG19, Xception and EfficientNet) are used to classify traffic signs using a state-of-art traffic signs dataset (German Traffic Sign Detection Benchmark-GTSDB). Accuracy was used as the classification evaluation criterion. The CNN model designed for the study gave the best result with an accuracy rate of 98% and a model competing with the literature was proposed.

YOLO-ADual: A Lightweight Traffic Sign Detection Model for a Mobile Driving System

Traffic sign detection plays a pivotal role in autonomous driving systems. The intricacy of the detection model necessitates high-performance hardware. Real-world traffic environments exhibit considerable variability and diversity, posing challenges for effective feature extraction by the model. Therefore, it is imperative to develop a detection model that is not only highly accurate but also lightweight. In this paper, we proposed YOLO-ADual, a novel lightweight model. Our method leverages the C3Dual and Adown lightweight modules as replacements for CPS and CBL modules in YOLOv5. The Adown module effectively mitigates feature loss during downsampling while reducing computational costs. Meanwhile, C3Dual optimizes the processing power for kernel feature extraction, enhancing computation efficiency while preserving network depth and feature extraction capability. Furthermore, the inclusion of the CBAM module enables the network to focus on salient information within the image, thus augmenting its feature representation capability. Our proposed algorithm achieves a mAP@0.5 of 70.1% while significantly reducing the number of parameters and computational requirements to 51.83% and 64.73% of the original model, respectively. Compared to various lightweight models, our approach demonstrates competitive performance in terms of both computational efficiency and accuracy.