As technology progresses, so does everything around us, such as televisions, mobile phones, and robots, which grow wiser. Of these technologies, artificial intelligence (AI) is used to aid the computer in making decisions comparable to humans, and this intelligence is supplied to the machine as a model. As AI deals with the concept of Black-Box, the model’s decisions were poorly comprehended by the end users. Explainable AI (XAI) is where humans can understand the judgments and decisions made by the AI. Earlier, the predictions made by the AI were not as easy as we know the data now, and there was some confusion regarding the predictions made by the AI. The intention for the use of XAI is to improve the user interface of products and services by helping them trust the decisions made by AI. The machine learning (ML) model White-box shows us the result that can be understood by the people in that domain, wherein the end users cannot understand the decisions. To further enhance traffic signal detection using XAI, the concept called local interpretable model- agnostic explanation (LIME) algorithm has been taken into consideration and the performance is improved in this paper.
1 August 2024 in International Journal of Informatics and Communication Technology (IJ-ICT)
As decades go by, technology advances and everything around us becomes smarter, such as televisions, mobile phones, robots, and so on. Artificial intelligence (AI) is applied in these technologies where AI assists the computer in making judgments like humans, and this intelligence is artificially fed to the model. The self-driving technique is a developing technology. Autonomous driving has been a broad and fast-expanding technology over the last decade. This model is carried out using the tiny you only look once (YOLO) algorithm. YOLO is mainly used for object detection classification. Tiny YOLO model is explored for the traffic signal detection. ROBI FLOW dataset is used for object detection which contains 2000+ image data to train the tiny YOLO model for traffic signal detection in real time. This model gives an improved accuracy and lightweight implementation compared to other models. Tiny YOLO is fast and accurate model for real-time traffic signal detection.
1 December 2023 in Indonesian Journal of Electrical Engineering and Computer Science
Lane detection is a crucial module for today’s autonomous driving cars. Detecting road lanes is a challenging task as it varies in color, texture, boundaries and markings. Traditional lane detection techniques detect the lane by applying a model trained with centralized data. As roads vary in urban and rural areas, a more localized and decentralized training technique is desired for accurate and personalized lane detection. Federated learning has recently proved to be a promising technology that trains and prunes the model using local data. Applying federated learning-based lane detection improves the accuracy of detection and also ensures the security and privacy of autonomous cars. This paper proposes FedLANE, a federated learning-based lane detection technique. U-Net, U-Net long short-term memory (LSTM) and AU-Net architectures were explored using a federated learning approach. Experimental analysis using TuSimple and CuLane dataset shows that the FedLANE based lane detection performs similar to that of the traditional deep learning lane detection models.
Face spoof detection and authentication systemhas significant role in IoT application that include validation of identity for computer security atm access crime detection and social care. Even though the authentication and classification system are vulnerable to different types of attacks. Presentation attack is a clear threat for facial and biometric based security and authentication applications. The discussed issues can be solved using artificial neural networks for face proof detection in IoT platform. The deep learning approaches for feature extraction in multicolor space is useful for obtaining more information from the input face regarding chrominance and luminance data. The extracted features can be selected and combined with minimum redundancy maximum relevance algorithm for providing discriminate and efficient feature set.
1 January 2024 in ITEGAM- Journal of Engineering and Technology for Industrial Applications (ITEGAM-JETIA)
Autonomous cars are revolutionizing transportation by navigating roadways without human intervention using digital technology and artificial intelligence. However, reliable lane recognition is a big barrier in this endeavor. Lane identification is a complex topic that presents significant challenges to computer vision and machine learning systems. Accurate lane line detection can be challenging due to real-world driving conditions, resulting in negatively impacts steering angle prediction. In response to this difficulty, our research proposes a novel strategy to improving lane detection and steering control accuracy. To recognize lanes with better precision, we use computer vision techniques, namely semantic segmentation. Semantic segmentation allows the vehicle's internal artificial intelligence system to classify each pixel in an image as belonging to a given object class, such as road lanes. The precise lane detection required for secure and dependable navigation is addressed by this suggested methodology, which addresses a crucial part of autonomous driving technology. In this paper we have improved the accuracy and robustness of autonomous vehicles, preparing them to face the difficulties of real-world road conditions, by using HR-Net architecture.