以影像辨識與分析為基礎之扁平足自動分類方法

The Flatfoot Automatic Classification Method Based on Image Recognition and Analysis

洗藝芬(Yi-Fen Hsi)

指導教授 ： 簡銘伸

國立虎尾科技大學/電機資訊學院/資訊工程研究所/碩士(2013年)

摘要

近年來由於醫療影像辨識與分析的進步，不論在理論方法或是量測設備上都有顯著的進展。藉由攝影機、成像設備產生的電子圖像，皆可以透過演算法進行功能性運算，進而判斷出細微的角度差距或分類出相關特徵。
本篇論文主要在給予一個簡單便利的操作介面，協助醫師在判斷足部問題上的速度及準確性，透過足部的X光影像，經由高斯平滑去除雜訊與邊緣偵測等影像處理後，將圖片分段討論，使用各區段合適的演算法找出判斷點來自動辨別是否為扁平足，並以清晰的線條標示出判斷依據及提供相關數據。另外在電腦分類之後也可經由醫師手動調整判斷點，提供醫師作為診斷的輔助工具，使得診療結果更加準確、客觀。

關鍵字

Medical Imaging ； Image Processing ； Gaussian Filter ； Edge Detection ； Flatfoot

並列摘要

In this thesis, the simple interface that help doctors determine the flat feet was proposed. Via using image recognition and analysis such as Gaussian filter and Canny edge detection, the proposed method can remove noise and processing X-ray images of the foot.
The individual X-ray image is divided into four blocks according to the proposed division method. According to the proposed algorithms, the system can find out four decision points of each block. Then, the system can automatically identify whether the flat feet is or not. The information and identification can be provided the doctor. In addition, the decision point can be also manually selected In other words, according to the selection made by the doctor, the system can make the results more accurately and objectively.

並列關鍵字

醫療影像 ； 影像處理 ； 高斯濾波 ； 邊緣偵測 ； 扁平足

參考文獻

[8] Pin-Ya Chou etc., ”The Influence of Different Imaging Postures on Flatfoot Evaluation”, 中華放射線技術學雜誌, 33, pp. 63-68, June 2009.

[2] Kalender, Willi A., Computed Tomography, Baker & Taylor Books, North Carolina, 2011.

[3] Buxton, Richard B., Introduction to Functional Magnetic Resonance Imaging, Cambridge Univ Pr, England, 2009.

[4] Neuman, M.R. etc., “Advances in Medical Devices and Medical Electronics”, Proceedings of the IEEE, 100, pp. 1537 – 1550, May 2012.

[5] Blankenberg, F.G., “Molecular imaging with single photon emission computed tomography”, Engineering in Medicine and Biology Magazine, IEEE, 23, pp. 51 – 57, July. 2004.