Abstract: Object recognition is the task of recognizing the object and labelling the object in an image or video scene. The proposed method seeks to implement an object recognition model for climatic data. Accurate characterization of objects in climate simulations and observational data archives is critical for understanding the trends and potential impacts of events in the climate. An object in an image can be recognized by extracting the features like colour, texture, or shape. Based on these features, objects of large-scale weather patterns are classified into various classes and each class is assigned a name. This paper presents an overview of object recognition methods by including two classes of object detectors. Two stage detectors such as Faster R-CNN focus more on accuracy, whereas the primary concern of one stage detectors such as YOLOv3 is speed. Faster Region-based Convolutional Neural Network method (Faster R-CNN) and You Only Look Once (YOLO) are the algorithms used for object recognition in this project. The results obtained from the two prominent approaches Faster R-CNN and YOLOv3 are compared
Keywords - Object recognition, climatic data, Convolutional neural network, Yolov3, Faster RCNN.
|
DOI:
10.17148/IJARCCE.2021.101115
