Faculty Publications
Permanent URI for this communityhttps://idr.nitk.ac.in/handle/123456789/18736
Publications by NITK Faculty
Browse
19 results
Search Results
Item An improved contextual information based approach for anomaly detection via adaptive inference for surveillance application(Springer Verlag service@springer.de, 2017) Rao, T.J.N.; Girish, G.N.; Rajan, J.Anomalous event detection is the foremost objective of a visual surveillance system. Using contextual information and probabilistic inference mechanisms is a recent trend in this direction. The proposed method is an improved version of the Spatio-Temporal Compositions (STC) concept, introduced earlier. Specific modifications are applied to STC method to reduce time complexity and improve the performance. The non-overlapping volume and ensemble formation employed reduce the iterations in codebook construction and probabilistic modeling steps. A simpler procedure for codebook construction has been proposed. A non-parametric probabilistic model and adaptive inference mechanisms to avoid the use of a single experimental threshold value are the other contributions. An additional feature such as event-driven high-resolution localization of unusual events is incorporated to aid in surveillance application. The proposed method produced promising results when compared to STC and other state-of-the-art approaches when experimented on seven standard datasets with simple/complex actions, in non-crowded/crowded environments. © Springer Science+Business Media Singapore 2017.Item Automated Parking System in Smart Campus Using Computer Vision Technique(Institute of Electrical and Electronics Engineers Inc., 2019) Banerjee, S.; Ashwin, T.S.; Guddeti, R.M.R.In today's world we need to maintain safety and security of the people around us. So we need to have a well connected surveillance system for keeping active information of various locations according to our needs. A real-time object detection is very important for many applications such as traffic monitoring, classroom monitoring, security rescue, and parking system. From past decade, Convolutional Neural Networks is evolved as a powerful models for recognizing images and videos and it is widely used in the computer vision related work for the best and most used approach for different problem scenario related to object detection and localization. In this work, we have proposed a deep convolutional network architecture to automate the parking system in smart campus with modified Single-shot Multibox Detector (SSD) approach. Further, we created our dataset to train and test the proposed computer vision technique. The experimental results demonstrated an accuracy of 71.2% for the created dataset. © 2019 IEEE.Item Optimized Object Detection Technique in Video Surveillance System Using Depth Images(Springer, 2020) Shahzad Alam, M.; Ashwin, T.S.; Guddeti, R.M.R.In real-time surveillance and intrusion detection, it is difficult to rely only on RGB image-based videos as the accuracy of detected object is low in the low light condition and if the video surveillance area is completely dark then the object will not be detected. Hence, in this paper, we propose a method which can increase the accuracy of object detection even in low light conditions. This paper also shows how the light intensity affects the probability of object detection in RGB, depth, and infrared images. The depth information is obtained from Kinect sensor and YOLO architecture is used to detect the object in real-time. We experimented the proposed method using real-time surveillance system which gave very promising results when applied on depth images which were taken in low light conditions. Further, in real-time object detection, we cannot apply object detection technique before applying any image preprocessing. So we investigated the depth image by which the accuracy of object detection can be improved without applying any image preprocessing. Experimental results demonstrated that depth image (96%) outperforms RGB image (48%) and infrared image (54%) in extreme low light conditions. © 2020, Springer Nature Singapore Pte Ltd.Item Automated Traffic Light Signal Violation Detection System Using Convolutional Neural Network(Springer, 2020) Bordia, B.; Nishanth, N.; Patel, S.; Anand Kumar, M.; Rudra, B.Automated traffic light violation detection system relies on the detection of traffic light color from the video captured with the CCTV camera, detection of the white safety line before the traffic signal and vehicles. Detection of the vehicles crossing traffic signals is generally done with the help of sensors which get triggered when the traffic signal turns red or yellow. Sometimes, these sensors get triggered even when the person crosses the line or some animal crossover or because of some bad weather that gives false results. In this paper, we present a software which will work on image processing and convolutional neural network to detect the traffic signals, vehicles and the white safety line present in front of the traffic signals. We present an efficient way to detect the white safety line in this paper combined with the detection of traffic lights trained on the Bosch dataset and vehicle detection using the TensorFlow object detection SSD model. © 2020, Springer Nature Singapore Pte Ltd.Item Hardware Accelerator for Object Detection using Tiny YOLO-v3(Institute of Electrical and Electronics Engineers Inc., 2021) Sharma, M.; Rahul, R.; Madhusudan, S.; Deepu, S.P.; Sumam David, S.For applications that require object detection to be performed in real-time, this paper presents a custom hardware accelerator, implementing state of the art Tiny YOLO-v3 algorithm. The proposed architecture achieves a reasonable tradeoff between the speed of computation (measured in frames per second or FPS) and the hardware resources required. Each CNN layer is pipelined and parameterized to make the complete design re-configurable. The proposed hardware accelerator was synthesized using the SCL(Semi-Conductor Laboratory, India) 180 nm CMOS process and also using Vivado Xilinx software with Virtex Ultrascale+ FPGA as the target device. The pipelined architecture, along with other architectural novelties, provided a higher frame-rate of 32.1 FPS and a performance of 166.4 GOPS at 200 MHz clock frequency. © 2021 IEEE.Item YOLOv5 Model-based Ship Detection in High Resolution SAR Images(Institute of Electrical and Electronics Engineers Inc., 2023) Sapna, S.; Sandhya, S.; Shetty, R.D.; Pais, S.M.; Bhattacharjee, S.Detection of ships in Synthetic Aperture Radar (SAR) images play a crucial role in maritime surveillance, most importantly under complex sea conditions. SAR permits observation in any weather conditions, at all hours of the day and night. At present, the ship detection from SAR images is a notable area of research since it is very difficult to detect the ships in the SAR images using traditional object or target detection algorithms. In this work, a You Only Look Once version 5 (YOLOv5) based ship detection model from SAR images with faster training speed and higher accuracy is implemented and tested. This model achieved a mean average precision (mAP) of 96.2% with a training time of 8.63 hours. This work also provides a comparative analysis with the existing methods for detection of ships in SAR images. The comparison shows that the YOLOv5 based model performs better in terms of both mean average precision and training time when compared to the existing models. © 2023 IEEE.Item Detection of Obstacles and Intelligent Narrow Road Navigation for Autonomous Vehicles - A Hardware Approach(Institute of Electrical and Electronics Engineers Inc., 2023) Gavasane, U.; Ananthanarayana, V.S.Obstacle detection is the most important functionality in the field of autonomous vehicles. One of the common challenges is detecting a narrow driving lane and navigating through it while performing obstacle identification and handling them. This problem falls under categories such as single-lane road block navigation, ad-hoc obstacles, narrow road path-finding, bridges, tunnel navigation, etc. In this work, we present an algorithm to handle these cases to perform successful detection and navigation through them. The lane navigation algorithm is used to perform autonomous driving under normal circumstances, and under narrow road circumstances, the Narrow Road Navigation (NRN) algorithm will be deployed. As proof of concept, we present a hardware prototype of a miniature car that will use the NRN algorithm. The hardware components and experimental results show that a successful implementation of the said algorithm is possible and can be used in actual real-life models and scenarios. © 2023 IEEE.Item Object detection in hyperspectral images(Elsevier Inc., 2022) Lone, Z.A.; Pais, A.R.Object Detection is a task of estimating and locating an object precisely in an image. It is a fundamental problem in computer vision and has been studied extensively in low dimensional images like RGB, grayscale, etc. High dimensional images like Hyperspectral images (HSI) contain ample information and are very powerful in enhancing the fine spectral differences between different objects. The advancement in spectral sensor technologies is making hyperspectral data more readily available, making it a promising technology for image analysis tasks. HSI has been explored in the fields of remote sensing, biomedical imaging, mineral classification, goods quality assessment, and object detection etc. The research concerning object detection in HSI has been gathering pace in recent times. This survey paper is an attempt to create a resource for researchers in the field. This paper provides a comprehensive review of both Supervised and Salient object detection. Moreover, a collection of important datasets is mentioned. We conclude the paper by mentioning research challenges and the future directions for the research in the field. © 2022 Elsevier Inc.Item Carotid wall segmentation in longitudinal ultrasound images using structured random forest(Elsevier Ltd, 2018) Yamanakkanavar, Y.; Asha, C.S.; Teja A, H.S.; Narasimhadhan, A.V.Edge detection is a primary image processing technique used for object detection, data extraction, and image segmentation. Recently, edge-based segmentation using structured classifiers has been receiving increasing attention. The intima media thickness (IMT) of the common carotid artery is mainly used as a primitive indicator for the development of cardiovascular disease. For efficient measurement of the IMT, we propose a fast edge-detection technique based on a structured random forest classifier. The accuracy of IMT measurement is degraded owing to the speckle noise found in carotid ultrasound images. To address this issue, we propose the use of a state-of-the-art denoising method to reduce the speckle noise, followed by an enhancement technique to increase the contrast. Furthermore, we present a novel approach for an automatic region of interest extraction in which a pre-trained structured random forest classifier algorithm is applied for quantifying the IMT. The proposed method exhibits IMTmean ± standard deviation of 0.66mm ± 0.14, which is closer to the ground truth value 0.67mm ± 0.15 as compared to the state-of-the-art techniques. © 2018 Elsevier LtdItem UAV based cost-effective real-time abnormal event detection using edge computing(Springer, 2019) Shahzad Alam, M.S.; Natesha, B.V.; Ashwin, T.S.; Guddeti, R.M.R.Recent advancements in computer vision led to the development of a real-time surveillance system which ensures the safety and security of the people in public places. An aerial surveillance system will be advantageous in this scenario using a platform like Unmanned Aerial Vehicle (UAV) will be very reliable and can be considered as a cost-effective option for this task. To make the system fully autonomous, we require real-time abnormal event detection. But, this is computationally complex and time-consuming due to the heavy load on the UAV, which affords limited processing and payload capacity. In this paper, we propose a cost-effective approach for aerial surveillance in which we move the large computation tasks to the cloud while keeping limited computation on-board UAV device using edge computing technique. Further, our proposed system will maintain the minimum communication between UAV and cloud. Thus it not only reduces the network traffic but also reduces the end-to-end delay. The proposed method is based on the state-of-the-art YOLO (You Only Look Once) technique for real-time object detection deployed on edge computing device using Intel neural compute stick Movidius VPU (Vision Processing Unit), and we applied abnormal event detection using motion influence map on the cloud. Experimental results demonstrate that the proposed system reduces the end-to-end delay. Further, Tiny YOLO is six times faster while processing the frames per second (fps) when compared to other state-of-the-art methods. © 2019, Springer Science+Business Media, LLC, part of Springer Nature.