An Application on Sweeping Machines Detection Using YOLOv5 Custom Object Detection

In the area of artificial intelligence and computer vision, there are two major categories: one being classification which interns to group set of items into their respective categories. Second is object detection which is localization of the object finding the x and y coordinates of a specific object. For a task-specific application, how to train and make a model for that specific job and use it as a standalone application. This paper shows you the entire process required to train and deploy the custom object detection model using yolo architecture. The processes consists of manual labeling, augmentation of data, configure files, and exporting files as weights. This method can be used for developing applications like marine life monitoring, traffic count finding, etc., we took sweeping machines as our objects to detect the time of arrival and average time spent on cleaning a particular area. By this, we can ensure that the cleaning operation is performed as per the schedule if not required actions will be took against them respectively.