透過您的圖書館登入
IP:3.128.173.32
  • 學位論文

利用手機為盲人偵測及指出特異移動物件

Detecting and Identifying a Distinct Moving Object with a Mobile Phone for the Visually Disabled People

指導教授 : 杜迪榕

摘要


視覺是人類的一個很重要的感官能力,但是視覺障礙者在這方面相當地薄弱,也使得視覺障礙者的生活產生極大的不便。在另一方面,視覺障礙者的聽覺常常比一般人還要敏銳。為了幫助視覺障礙者並改進其生活品質,進而使他們提早知道危險的存在,我們實作出一個以聽覺代替視覺的可攜式系統來幫助視覺障礙者。在這個研究中,為了讓視覺障礙者便於攜帶,硬體本身必須具有體積小且重量輕之特性,所以我們將發展平台設定在手機上的原因是體積小重量輕容易攜帶而且價格不貴。當然,所選用之手機必須能拍照及輸出立體聲,才能以立體聲音展示所拍攝之物件。另外,我們也提出了一個方法,可以偵測移動的物件,藉此視覺障礙者可以大致知道移動物件的外觀,甚至可以辨識出移動物件。視障者可以藉由系統所發出的立體聲音,來判斷出移動物件的位置。因為人類的耳朵解析度及反應速度非常有限,所以我們的研究增加了偵測柱子和樓梯的功能,未來還要增加其他危險物的偵測。因為我們的系統是建立在JAVA的技術上,所以任何一種可執行JAVA的裝置,如手機或PDA,只要具有照相及立體聲之功能皆可是我們系統之平台。

並列摘要


Vision is a very important sense for mankind, but the visually disabled people are lack of this sense, which is very inconvenient for their life. On the other hand, the visually disabled people have sharper hearing sense than normal people. Therefore, an embedded system for constructing scenes by hearing is implemented to improve the life of the visually disabled people and the embedded system can let the visually disabled people detect danger in their environment. The hardware of the system must be small and light enough such that the visually disabled people can take it easily. Therefore, the embedded system is made by a mobile device because it is inexpensive, small and light. The adopted mobile devices have a software controllable camera and a stereo audio device for taking pictures and outputting stereo sound to show scenes and moving objects. Additionally, a simple method to detect a distinct moving object in a scene is proposed. Therefore, based on the developed system, the visually disabled people can hear the sketches of the distinct moving object and even recognize what the object is. The visually disabled people can be noticed where the location of the moving object is via a stereo sound generated from the mobile device. Since the resolution and sensitivity of human hearing is limited, the functions of detecting pillars and stairs are also developed and some functions of deteting other dangerous objects will be included in the near future. Because our system is set up based on JAVA technology, any JAVA-enabled device such as cell phone or PDA, which equips with a camera and a stereo audio device, can be the platform of our system.

參考文獻


[1] A. Censi, A. Fusiello, and V. Roberto, “Image stabilization by features tracking.” Proc. the 10th Int. Conf. on Image Analysis and Processing (ICIAP ’99), pp. 665-667, Sept. 1999.
[2] L. Dreschler and H.-H. Nagel, “Volumetric model and 3D trajectory of a moving car derived from monocular TV frame sequences of a street scene.” Proc. of Int. Joint Conf. on Artificial Intelligence, Vancouver, Canada, pp. 692-697, 1981.
[3] D.-R. Duh, C.-W. Lu, and K.-H. Wang, “Constructing scenes by hearing of the visual disabilities,” Proc. National Computer Symp., Yung-Kang City, Tainan Hsien, Taiwan, p. 97, Dec. 15-16, 2005.
[4] D.-R. Duh, J.-X. Wu, L.-C. Yang, and J.-S. Jiang, “Detecting and identifying a distinct moving object with a mobile phone for visually disabled people,” Proc. 4th Int. Conf. in Information Technology and Application, Harbin, China, vol. 1, pp. 248-253, Jan. 15-18, 2007.
[5] G. L. Foresti and C. Micheloni, “A robust feature tracker for active surveillance of outdoor scenes,” Electronic Letters on Computer Vision and Image Analysis, vol. 1, no. 1, pp. 21-34, 2003.

延伸閱讀