Artificial Intelligence Techniques for Identifying and Detecting Objects

Abstract

AI has become necessary due to the use of deep learning and machine learning techniques. AI seeks to reduce human intervention in order to automate tasks. IT, healthcare, banking, and agriculture all make extensive use of it. Several deep learning algorithms that mimic the intelligence of the human brain are used to do this. It is possible to adjust these AI algorithms in accordance with evolving requirements and increased effectiveness. In order to classify the photographs and identify the things they include, this research attempts to make use of the advancements in AI technology. CNN (Convolutional Neural Networks) is a popular AI method. The CNN is a multi-layered deep learning system that extracts and filters the parameters found in the pictures. To increase the accuracy of picture recognition, certain extra layers of the CNN algorithm and ResNet50 are utilized to extract the parameters. ImageNet is the picture dataset used to train and evaluate the suggested model. Prior to being sent to the suggested model, the photos are first processed. The photos retrieved following the first processing are used to train, validate, and test the suggested model. Until the highest level of precision is achieved, the same procedure is carried out several times. It is noted how well the suggested model is in recognizing images. A comparison is made between the results achieved and various picture classification methods, such as VGG16 and VGG19. In terms of accuracy, it is determined that the suggested model performs better than other conventional techniques.