Compressed Deep Learning Model for Detecting COVID-19 Disease: A Genetic Algorithm based approach

A recent pandemic of Covid 19 proves a black spot of our medical industry and it also shows us importance of computer vision to detect the Covid-19 in early stage. It is very important to detect Covid 19 early to prevent the rapid spread of the virus. In application of computer vision we have deep learning (DL) based model to detect the covid 19 and in DL model required wide range of data to process for training the model. Data processing and analysis are becoming more difficult because of the development of Internet of Things (IoT) devices and cameras, which has resulted in increased data generation. Computer vision along with deep learning, a branch of artificial intelligence, has become a critical tool for processing these gadgets’ massive volumes of data. As a result, developments such as intelligent houses, driverless vehicles, and surveillance systems have been created. However, the complexity of deep learning models and the data required for training can lead to high computational costs and large model sizes. This has led to the need for model compression techniques to reduce the size of deep learning models without sacrificing performance. However, substantial computing costs and big model sizes might arise from the complexity of deep learning models and the volume of data needed for training. Using compression methods, these models can be made smaller without compromising accuracy. In conclusion, the growth of IoT devices and cameras has increased the demand for deep learning methods of data analysis. Deep learning models can be more effective using model reduction, making them better suited for Internet of Things (IoT) devices with constrained storage capacity and processing. In this article, we have proposed a compressed deep learning model using genetic algorithm for covid 19 detection. According to our experiment, the precision of the proposed model is 96. 94% where the compressed CNN size is less than the original CNN.