• Bhagyashri S. Chandekar
• Hirendra R. Hajare

Object detection aims to acknowledge all instances of a known class of objects in a picture, like people, vehicles, or faces. In recent times deep learning techniques was applied for detecting objects and people existing methods have some problems with viewpoint changes and occlusion. This paper proposes an answer for automatic object detection by implementing instance segmentation at a pixel level, and a number of other RCNN techniques were also inferred during this paper. The proposed Mask RCNN detects the photographs and marks them with bounding boxes, class labels still as masks. The implemented mask RCNN model was trained and tested with COCO dataset in order that the model is capable to detect objects in an exceedingly congested image. The model was also tested during a custom dataset to test the model performance, and therefore the model obtained 94% mAP for the custom dataset. This paper also mentions the long run scope of this work in order that the robustness and also the reliability of the model may be improved further.

Computer vision; CNN; Deep learning; Object detection; Mask RCNN; Fast R-CNN

"Deep Learning based Object Detection using Mask RCNN", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.9, Issue 5, page no.k531-k537, May-2022, Available :http://www.jetir.org/papers/JETIR2205B71.pdf

"Deep Learning based Object Detection using Mask RCNN", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.9, Issue 5, page no. ppk531-k537, May-2022, Available at : http://www.jetir.org/papers/JETIR2205B71.pdf