影像處理技術於果蠅活動能力監測之研究

Image Processing Study on Monitoring Activity of Drosophila

吳鈺徵(Yu-Cheng Wu)

指導教授 ： 蔡宏營

國立清華大學/工學院/動力機械工程學系/碩士(2012年)

摘要

由於果蠅的基因與人類有60%的相似度，且果蠅各種行為與人類相似，諸如：睡眠行為、求偶行為等。果蠅亦具有生長周期短、基因轉植技術成熟的優點，尤其是人腦的神經結構複雜，果蠅腦神經結構相對於人腦較為簡單。因此，果蠅在行為學以及腦神經科學領域被大量的運作為研究的對象。而過去研究中多利用人力方法進行數據的分析，此方法容易造成誤差以及浪費人力資源與時間。
　　本研究目的在於運用影像方法針對果蠅睡眠行為與果蠅對氣味產生的行為開發程式以進行分析，期望藉此達到減少人力資源與時間的浪費。為研究果蠅於平台內的行為需要先將果蠅由影像中抓取出來，使用影像處理方法中的瞬間差異法與背景消去法進行移動物的擷取，其中為了取得最好之背景影像運用多張影格合成，加以運用二值化方法與形態學方法去除雜訊以及將相連之個體分離。在睡眠行為研究上開發出可以同時偵測果蠅運動、休息與睡眠三種數據之程式，並運用統計方法定義出休息狀態之參數。此外，運用影像處理方法模擬現行睡眠實驗使用之Drosophila Activity Monitoring System(DAMS)，驗證針對本研究所使用之影片其影像處理方法有較正確的分析結果。而在果蠅對氣味所產生的行為分析軟體上可以針對多隻果蠅在平台內之分布情形進行統計，並經由實驗驗證此自動化系統與分析程式之成效。藉由這些行為研究程式的開發，有助於有效提高實驗的效率，並且取得最正確之實驗數據。

關鍵字

影像處理 ； 果蠅 ； 活動能力

並列摘要

Since Drosophila’s genes are 60% similar to human beings’, behaviors of Drosophila and human beings’ are alike such as sleep and courtship. Furthermore, the short life of Drosophila and the matured transgenic techniques make Drosophilae widely used in biological experiments, behavior research and neuroscience. However, most of experiments depend on labor to analyze the results. It is time-wasting and inefficient. In addition, data errors can be a big issue between experiments by different researchers.
In the current study, image processing methods have been developed for monitoring the sleeping behavior of a Drosophila group and the behavior affected by different odors in order to increase the analysis efficiency and the correctness. In both algorithms, the frame difference method is used to capture the moving object/Drosophilae in the recorded video. Binarization and morphology methods are used to eliminate the noises. In addition, the overlapped Drosophila can be identified based on the morphology method. In the current sleeping calculator, the sleep period, the sleep bout number, the sleep bout length, the activity period, the activity bout number, the activity bout length, the rest period, the rest bout number, the rest bout length and the speed can be obtained. Besides, three states of Drosophila are redefined for better meaning. In addition, the image processing method is used to simulate the Drosophila Activity Monitoring System (DAMS), which is widely used for sleeping behavior of Drosophila in recent study. Image processing helps to increase the higher accuracy than DAMS does. The objective of odor algorithm is to develop a monitoring system which includes the air/odor flow equipment to study the behavior of Drosophila affected by different odors. With the help of these algorithms, higher experimented efficiency, accuracy and correctness can be provided.

並列關鍵字

Image Processing ； Drosophila ； Activity

參考文獻

[1] D. Gavrila and S. Munder, “Multi-cue pedestrian detection and tracking from a moving vehicle,” International Journal of Computer Vision, vol. 73, pp. 41-59, 2007.

[2] S. Zehang, G. Bebis and R. Miller, “On-road vehicle detection: a review,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol.28, pp.694-711, 2006.

[3] T. B. Moeslund, A. Hilton and V. Krüger, “A survey of advances in vision-based human motion capture and analysis,” Comput. Vis. Image Underst., vol. 104, pp. 90-126, 2006.

[5] B. K. P. Horn and B. G. Schunck, “Determining optical flow,” Artificial Intelligence, vol. 17, pp. 185-203, 1981.

[8] S. Y. Elhabian, K. M. E. Sayed and S. H. Ahmed, “Moving object detection in spatial domain using background removal techniques-State-of-Art,” Recent Patents on Computer Science, vol. 1, pp. 32-54, 2008.