Table 6.

Comparison with state-of-the-art methods.

Method	Components	Dataset	Result	Coverage Area	Connection	Cost
Hoang et al. [41]	Mobile Kinect, laptop Electrode matrix, headphone and RF transmitter	Local dataset	Detect obstacle and generate audio warning	Indoor	Offline	High
Bai et al. [8]	Depth camera, glasses, CPU, headphone and ultrasonic sensor	Not included	Obstacle Recognition and audio output	Indoor	Offline	High
Yang et al. [42]	Depth Camera on Smart glass, Laptop, and headphone	ADE20, PASCAL, and COCO	Obstacle Recognition and generate clarinet sound as warning	Indoor, Outdoor	Internet Required	High
Mancini et al. [43]	Camera, PCB, and vibration motor	Not included	Obstacle recognition and vibration feedback for the direction	Outdoor	Offline	Low
Bauer et al. [44]	Camera, smartwatch, and smartphone	PASCAL VOC Dataset	Object detection with direction of object into audio output	Outdoor	Internet Required	High
Patil et al. [45]	Sensors, vibration motors,	No Dataset	Obstacle detection with audio output	Indoor, Outdoor	Offline	Low
Eckert et al. [46]	RGB-D camera and IMU sensors	PASCAL VOC dataset	Object detection with audio output	Indoor	Internet Required	High
Parikh et al. [47]	Smartphone, server, and headphone	Local dataset of 11 objects	Object detection with audio output	Outdoor	Internet Required	High
AL-Madani et al. [48]	BLE fingerprint, fuzzy logic	Not included	Localization of the person in the building	Indoor	Offline (Choice Wi-Fi or BLE)	Low
Proposed Method	RGB Camera, Distance Sensor, DSP processor, Headphone	Local dataset of highly relevant objects for VIP	Object detection, Count of objects, obstacle warnings, read text, and works in different modes	Indoor, Outdoor	Offline	Low