http://localhost:8080/xmlui/handle/123456789/7375 Mon, 13 Apr 2026 00:15:50 GMT 2026-04-13T00:15:50Z http://localhost:8080/xmlui/handle/123456789/9832 Feature Extraction Techniques For Detection Of Rule Violation(S) On Highway At Toll Centres Using Raspberry Pi Hardware Purohit, Manishkumar NA For Full Thesis Kindly contact to respective Library Tue, 01 Sep 2020 00:00:00 GMT http://localhost:8080/xmlui/handle/123456789/9832 2020-09-01T00:00:00Z http://localhost:8080/xmlui/handle/123456789/9831 An Efficent Load Balancing Approach in Cloud Computing Shah, Jaimeel NA For Full Thesis Kindly contact to respective Library Fri, 01 May 2020 00:00:00 GMT http://localhost:8080/xmlui/handle/123456789/9831 2020-05-01T00:00:00Z http://localhost:8080/xmlui/handle/123456789/9820 An Efficient Deep Convolutional Neural Network Approach for Object Detection and Recognition from a Video Sequence using Multi-Scale Anchor Box Garg, Dweepna Deep learning is a new era of machine learning which trains computers to find the structure from a massive amount of data. Learning is described at multiple levels of representation. This enables us to make sense of the data consisting of text, sound, and images. Many computer vision problems such as object detection, image classification, and semantic segmentation have been solved using convolutional neural networks. Object detection in videos involves confirming the presence of the object in the image or video and then locating it accurately for recognition. Detecting and recognizing the still object from an image has comparatively shown better performance with the use of detection frameworks like R-CNN, Fast R-CNN, Faster R-CNN etc. The challenge is to detect and recognize the moving object from a static camera efficiently and accurately. In the video, modeling techniques suffer from high computation and memory costs which may lead to a decrease in performance measures such as accuracy and efficiency to identify the object accurately. The motive behind this work is to accurately detect and recognize the moving and still object from a video sequence using deep learning in real-time. The existing algorithms of object detection based on the deep convolution neural network worked well for large-size objects as the detection models get better results. However, those models fail to detect the varying size of the objects that have low resolution. This is because the features do not fully represent the essential characteristics of the objects in real-time after going through the repeated convolution operations of existing models. The proposed work improves the accuracy of detection by extracting the features of object at different size and scale by using a multi-scale anchor box. With the help of CNN, the deep knowledge from the dataset is extracted by giving the model a rigorous set of training samples. Our model has achieved 84.49 mAP on the test set of the Pascal VOC-2007 dataset which is higher than the state-of-the-art models. In our work, considering the accuracy as one of the evaluation measures, the objects get detected and recognized at 11 FPS which is comparatively better than other real-time object detection models. Our model is also trained and tested for face detection using the FDDB dataset. Moreover, the model is also able to detect partially covered faces. This also serves as one of the real-time application of our proposed work. For Full Thesis Kindly contact to respective Library Wed, 01 Jan 2020 00:00:00 GMT http://localhost:8080/xmlui/handle/123456789/9820 2020-01-01T00:00:00Z http://localhost:8080/xmlui/handle/123456789/9816 NXTGeUH: Internet of Medical Things Based Next Generation Ubiquitous Real-Time Health Monitoring System Patel, Warishkumar Background: The biggest challenge in this technologically advanced society is the improvement in the health of aging individuals. The focal cause for significant injuries and early death in senior citizens is due to falling, the possibility to automatically detect falls has increased demand for such devices, the high detection rate is achieved using the wearable sensors, this system has a quiet social and monetary impact on society. So even for the day-to-day activity in the life of aged people an automatically fall detecting systems and vital signs examining system becomes a necessity. We have added one extended idea (which is not in the scope of thesis) in this research by considering current COVID-19 outbreak. As proposed system can monitor vital signs of the patient or individual remotely in real-time without person to person contact. So our health workers, doctors, nurses can easily monitor vital signs of corona positive or infected patient remotely and get updated in real-time. So They may have less chances to get infection from positive one. So it is fact that in this era, IoMT based healthcare systems are highly needed to serve in such bigger outbreak and when country have very less resources in terms of Doctors, nurses, hospitals and medical equipment. As coronavirus keeps on spreading, specialists’ doctors and healthcare systems frameworks are confronting a large number of difficulties at all phases of the pandemic Objective: This research work aims at helping aged people and every other necessary human by monitoring their vital signs and fall prediction. A fall detecting VitalFall gadget which could analyze the measurement in all three orthogonal directions using a triple-axis accelerometer and Vital Signs Parameters (Heartrate, Heartbeat, and Temperature monitoring) for the ancient people with a Next-Generation Ubiquitous Healthcare Monitoring (NXTGeUH) approach with proposed VitaFall wearable device is proposed which is well-timed and gives an effective decision of the fall. The minimum value to define the probability of an old individual’s fall is evaluated by calculating the spur and gradient which people make with the parallel plane are with the Vital Signs Parameter, MPU6050 is a Tri-Axial Accelerometer and Tri-Axial Gyroscope Module and collects the accelerations as well as the (angular velocity) angle developed between the aged and the parallel plane of aged people for a VitalFall device in the Internet of Medical Things. A guardian can be notified by sending a text message via GSM and GPRS module in order that aged can be helped, however, a delay in the time is noticed when comparing the gradient and minimum value to predetermine the state of the old person. It is the era of IoT and ambient intelligence. There a greater number of serious problems in the older populace because of the hasty enlargement of modern society. Methods: Comparison with Present Algorithms there are various benefits regarding privacy, success rate and design of using an implemented algorithm over the existing algorithms assessed using Kappa analysis, Recall, Precision, Accuracy, and F1- Score.As concluded from the experimental outcomes. The NXTGeUH proposed system has succeeded to achieve 96.43%Accuracy, 94.06%Precision, 94.62% Recall, 94% F1-Score, when detecting falls. The proposed advanced algorithm NXTGeUH monitor’s the patient's count using proposed VitaFALL device with combining the decision of Fall assessment and Vital signs monitoring. For Full Thesis Kindly contact to respective Library Wed, 01 Jan 2020 00:00:00 GMT http://localhost:8080/xmlui/handle/123456789/9816 2020-01-01T00:00:00Z