Deer Hunting Optimization with Path Augmented RetinaNet Model for Scale-Invariant Object Detection and Tracking

In the zone of pattern recognition and machine learning, object detection and tracking are gaining interest because of its range of applications like visual surveillance. Recently, several deep learning-related techniques are provided for object classification however, there exists some set of concerns or problems that minimizes the overall classifier accuracy. Congest conditions, complex backgrounds, and similarities among various objects become some challenging problems. Large scale variation over object samples, and particularly, the task of detecting small objects turns out to be one such factor for the performance differences. This study introduces an Artificial Intelligence Enabled Scale Invariant Object Detection and Tracking (AIE-SIODC) model on Color Images. The presented AIE-SIODC model intends to recognize and track the objects exist in the color images irrespective of scaling. It follows a two stage process like object detection and objects classification. Initially, the presented AIE-SIODC technique applies path augmented RetinaNet (PA-RetinaNet) based object detection module, which improves the feature extraction process. To improvise the network potentiality of the PA-RetinaNet method, the deer hunting optimization (DHO) algorithm is utilized. Finally, extreme gradient boosting (XGBoost) classifier is exploited for classification procedures. For examining the effectual object detection outcomes of the AIE-SIODC technique, a wide ranging experimental validation process can be performed. The simulation values confirmed the enhancements of the AIE-SIODC method over other DL models.