Apparatus and method for identifying surrounding environment by means of image processing and for outputting the results

• US 20050275718A1
• Filed: 08/27/2004
• Published: 12/15/2005
• Est. Priority Date: 06/11/2004
• Status: Active Grant

• Current Assignee: Far Eastern Memorial Hospital, Oriental Institute Of Technology
• Original Assignee: Oriental Institute of Technology and Far Eastern Memorial Hospital
• Inventors: Chu, Yiu-Tong, Lun Lai, Chin
• Granted Patent: US 7,230,538 B2
• Time in Patent Office: Days
• US Class Current: 348/113
• CPC Class Codes: A61H 2003/063 with tactile perception
• CPC Class Codes: A61H 3/061 with electronic detecting or guiding means
• CPC Class Codes: G06T 7/73 using feature-based methods
• CPC Class Codes: G06V 20/584 of vehicle lights or traffic lights
• CPC Class Codes: H04N 7/183 for receiving images from a single remote source

Abstract

The present invention relates to an apparatus for identifying surrounding environment by means of image processing, so as to be used by blind people or unmanned vehicles. The said apparatus comprises an input device, which includes at least one CCD or CMOS means for capturing images, a processing device, which could be a micro processing unit such as a notebook computer, DSP or other embedded system, and an output device, such as a speech earphone. The processing device applies hue analysis and geometric analysis to identify traffic signals and markings.

Claims

1. An electronic apparatus for detecting and determining the environment, comprising:

an input device comprising at least one CCD or CMOS means so as to take images;

a processing device, which may be composed of a portable embedded processing system; and

an output device, which may be a speech notification device or tactual output device;

wherein the input device detects surrounding environment and transmit detected information to the processing device for determination, then the result of determination are output from the processing device to the output device.

2. The apparatus according to claim 1, wherein the input device includes an active detecting means such as a radar means or an ultrasonic means.

3. The apparatus according to claim 1, wherein the input device captures light with frequency both beyond and within the range of visible light.

4. The apparatus according to claim 1, wherein the processing device can determine traffic lights, traffic markings, and lines according to the information taken by the input device.

5. The apparatus according to claim 1, wherein the processing device can determine obstacles, moving objects and navigating information (such as determination of sidewalk edges, tactile tiles, ground level, etc.) according to the information taken by the input device.

6. The apparatus according to claim 1, wherein the processing device utilizes hue analysis and geometry analysis to determine traffic lights, traffic markings, and lines.

7. The apparatus according to claim 6, wherein the processing device determines crosswalks and its length, width, direction by geometry (straight line) detecting technique.

8. The apparatus according to claim 4 or 5, wherein the processing device can use techniques such as image differential analysis, binocular vision analysis, texture analysis, or frequency analysis to determine surrounding environment.

9. The apparatus according to claim 1, wherein the processing device can determine horizon by analyzing geometry and hue information in images taken by input devices so as to determine elevation angle of the input device.

11. The apparatus according to claim 1, wherein the processing device can include a system setup mode.

14. The apparatus according to claim 1, wherein the processing device can take the film recorded by the input device as the foundation of self-learning.

10. The apparatus according to claim 10, wherein the processing device can actuate the input device to adjust the elevation angle automatically, or actuate the output device to notify the user to adjust the elevation angle by hand.

12. The apparatus according to claim 12, wherein the system setup mode can be accessed by directly operating the processing device or by inputting predetermined signal into the input device.

13. The apparatus according to claim 12 or 13, wherein the system setup mode can include a self-learning function.

15. A method for identifying traffic lights comprising the following steps:

transforming color images taken by an input device into HSV format;

choosing areas with hue value close to traffic lights and consecutive to each other;

determining if the areas chosen are in the shape of the traffic lights; and

checking other auxiliary information such as;

if the light changes color between red, green, yellow, red in a reasonable time interval and within a limited area, if any auxiliary signals nearby, etc.

16. The method according to claim 15, wherein an identifying operation will restart from the first step if no areas with hue value close to traffic lights is found.

17. The method according to claim 15, wherein an identifying operation will restart from the first step if areas chosen are not in the shape of the traffic lights.