Vision Navigator: A Smart and Intelligent Obstacle Recognition Model for Visually Impaired Users

Table 2

Accuracy rate analysis of observed obstacles with varying obstacle detection models in indoor settings.

Environment	Model	Obstacle	Accuracy (%)	Mean Accuracy (%)
Indoor	Retina Net	Cat	92.5	93.62
		Human-1	93.2
		Table-1	92.6
		Door	94.6
		Chair-1	95.2
Indoor	Yolo Tiny	Dog	91.7	92.26
		Human-2	90.4
		Sofa	92.2
		Door-2	94.1
		Chair-2	92.9
Indoor	R-CNN	Human-3	91.2	91.48
		Bag	90.4
		Dustbin	89.3
		Board	93.7
		Table-2	92.8
Indoor	SSD-RNN	Chair-1	95.6	95.06
		Sofa	95.9
		Human-4	93.9
		Clothes	95.1
		Human-5	94.8