The document presents an algorithm for detecting and recognizing restroom signage in images. It proposes using shape detection to identify the head and body of signage, and then using SIFT features to match corners and recognize the specific signage template. An experiment on 102 images achieved 89.2% detection and 84.3% recognition rates. Challenges included view angle changes and complex backgrounds. The work was supported by NIH and NSF grants.
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
Camera-based Signage Detection and Recognition for Blind Persons
1. Shuihua Wang and Yingli Tian
{swang15, ytian}@ccny.cuny.edu
Presented by: Shizhi Chen
Department of Electrical Engineering
The City College of New York
12. Experiment Results
89.2% detection rate (91 out of 102 images)
84.3% recognition rate (86 out of 102 images)
Confusion matrix: column is the ground truth
13. Intermediate Results
Original
Image
Binary
Image
Connected
Component
W D M D
W M
Signage
Recognition
18. Author Contact
Shuihua Wang and Yingli Tian
{swang15, ytian}@ccny.cuny.edu
Hinweis der Redaktion
The detection of the RFID tag is done by the RFID system it self with the reader and the antenna. The human action part or once a bottle tag is removed from the range of the anteena, computer vision tools plays in roll. Algorithms like Background subtraction and color modeling are used to detect those moving bottle tags ONLY.Since the computation time is very long, we used multi thread programming in order to reduce computation time. Also as for the color detection, the HSV color model is used to detect the tags after the background subtraction model has done its job.We use HSV, because HSV describes color using more familiar comparisons such as color, vibrancy and brightness. And the tag color is just only pure Blue…