The discipline of computer vision makes extensive use of object detection, which is essential for many different applications. The approach to object recognition at this time has essentially grown into two categories: deep learning and classic machine learning approaches using a variety of computer vision techniques. Object detection methods are reviewed in this article. Initially, a summary and introduction of the current machine learning approaches are provided. In everyday life, it is usual to use video cameras to monitor the campus. The majority of these surveillance methods employ people to keep an eye on what is happening in the target region. Yet, employing humans for monitoring has drawbacks of its own. To get around this restriction, researchers are developing automated visual surveillance systems. Environment modelling, motion segmentation, object categorization, tracking, behavior comprehension, person identification, and data fusion are the phases that make up the visual surveillance process. Finding moving objects in a video sequence is the first and most important step in visual surveillance. A person, a car, or another item may be the moving object of interest. The technology known as object detection is concerned with determining the semantic class of the moving item in the video sequence. Hence, Object Detection is crucial for following moving objects and analyzing their behavior in the provided video sequence. This study discusses the many object detection techniques that are available, taking into account the significance of object detection in visual surveillance

#