Intelligent System for Social Distance Monitoring with Human Detection and Tracking using YOLOv3

Everyday life and the global economy have been negatively impacted by COVID-19 (Coronavirus). Slowing the spread of coronaviruses through social distance is proven to be an effective strategy in the war against COVID-19. The social distancing is the best way to stop the spread of COVID-19, as it prevents people from coming into intimate touch with each other. Recently, due to the fast spreading outbreak of the COVID-19, one of the obligatory preventive measures to avoid physical contact has become social distance. Surveillance methods that use Deep Learning, Open-CV and Computer vision to follow pedestrians and prevent congestion are the focus of this article. Closed-circuit television (CCTV) and drones can be used for implementation, where the camera will use object detection to identify the crowd and compute the distance between the humans. Local law enforcement will be notified if the Euclidean distance between two persons is less than the standard distance, which is determined by converting it to pixels and comparing it to that value.

Pneumonia_Net Model: Deep Learning Based Intelligent System forPneumonia Classification using Chest X-ray Images

Radiologists are faced with a challenging problem whenever they have to classify the anomalies shown on chest x-rays. Because of this, throughout the course of the last few decades, computer aided diagnostic (CAD) systems have been created to extract meaningful information from X-rays in order to assist medical professionals in gaining a quantitative understanding of an X-ray.Because radiology is such an important field, most of the time the analysis of radiologist images is carried out by trained medical professionals. This is due to the fact that patients seek the highest possible level of treatment in addition to the highest possible quality, regardless of how much it costs.However, its complexity and the subjective nature of the visuals limit its usefulness. There is a great deal of diversity between different translators and a great deal of exhaustion in human professional image processing. Our main goal is to classify lung disorders utilizing diagnostic X-ray images analysed using deep learning and images exploited using Pandas, Keras, Open CV, Tensor Flow, etc. Chest radiographs are still diagnosed by doctors and radiologists using manual and visual methods. As a result, a system capable of diagnosing chest X-rays must be developed that is both smart and automated. The goal of this study is to classify chest X-ray images into normal and pathological using a deep neural network model called Pneumonia Net. It is trained and evaluated using chest X-rays taken from publicly available databases that include both normal and pathological radiographs. Due to their capacity to automatically extract high-level representations from large data sets, CNN-based deep learning categorization approaches outperform existing picture classification methods in this regard. Three different network models are compared depending on their performance. In experiments, it was found that the Pneumonia Net model had a good generalisation capacity in identifying unseen chest X-rays as normal or anomalous, and that its performance was better than that of other network models.