植物型態觀測系統

Morphology and Morphometric Platform for Plant

賴峻廷(Jiun-Ting Lai)

指導教授 ： 蔡哲民

崑山科技大學/資訊科技學院/資訊管理研究所/碩士(2008年)

https://doi.org/10.6828/KSU.2008.00017

摘要

本論文利用影像處理中之物件切割與追蹤技術結合遠端遙控硬體之方式，提出一套具備巨觀及微觀功能之植物型態觀測系統。在巨觀部份，本論文運用改良之動態輪廓模型（Active Contour Model）演算法來製作一套具即時植物影像追蹤與生長量測記錄之觀測平台，此一平台使用CCD Camera進行影像的擷取，並以Java程式開發而成，使用者僅需以滑鼠針對影像中物件作概略的圈選，系統將利用改良後之動態輪廓模型演算法進行物件輪廓的收斂，並於判斷收斂完成後計算並記錄物件質心座標與輪廓區域面積，針對植物生長過程作長時間之自動觀測。在微觀部份，是針對使用者所採集之植物製成的切片標本至於顯微鏡下觀測，該微觀平台主要具有遠端線上多人標示討論與切片資料維護的功能，系統操作者可藉由網際網路遠距觀測實驗室中之切片，並可遙控顯微鏡作切片的移動與顯示倍率的切換等功能，遠端標示討論功能可作為一般生物相關課程之遠距教學平台，而標本切片資料維護功能則可提供生物研究時，使用者儲存切片資訊之用。本論文所製作之巨觀與微觀平台整合之資訊系統將有利於植物學研究時對於特有物種的觀測及培育條件之評估。

關鍵字

影像處理 ； 動態輪廓模型 ； 遠距教學

並列摘要

In this thesis, we introduce a plant observation system which can be used to study plants from macroscopic and microscopic views of the plants. In the macroscopic part, the system can track and measure the plant growth in real time from the images of the CCD camera. In the microscope part, the system can observe the slices of plants and other creature using the microscope images. In order to take the advantages of ubiquitous internet, this system is developed in a web-based style. Therefore, this system can be operated remotely. In addition, multiple users can observe and discuss remotely.
In order to correctly encircle the interested part or region in the image, we propose a modified active contour model which imposes the biological morphology characteristic to confine the snake movement and to avoid fault edge matching. The images, markings and the discussing result can be stored in the database and retrieved for later use. In order to precisely control the platform, we also develop a circuit board based on MSP430 micro-processor, to control the step motor of the stage.
With the ability of remote observation and control, this system can be applied to research in diverse areas of plants and other creatures. With the rich contents of database, this system can also be utilized in problem-based learning.

並列關鍵字

Image Processing ； Active Contour Model ； Distant Learning

參考文獻

[1] Y.T. Chi, C.F. Chien and T.T. Lin, “Leaf shape modeling and analysis using geometric descriptors derived from Bezier curves” Trans. of the ASAE, Vol. 46, No. 1, 2003, pp-175-185.

[2] C.F. Chien and T.T. Lin, “None-destructive growth measurement of selected vegetable seedlings using orthogonal images” Trans. of the ASAE, Vol. 48, No. 5, 2005, pp.1953-1961.

[6] Philippe Thévenaz and Michael Unser, “User-Friendly Semiautomated Assembly of Accurate Image Mosaics in Microscopy” Microscopy Research and Technique, Vol. 70, No. 2, February 2007, pp.135-146.

[7] Schmitt O., Eggers R. and Modersitzki J., “Videomicroscopy, Image Processing, and Analysis of Whole Histologic Sections of the Human Brain” Microscopy Research and Technique, Vol. 66, No. 4, 2005, pp.203-218.

[10] Nilanjan Ray, et al., “Merging Parametric Active Contours within Homogeneous Image Regions for MRI-Based Lung Segmentation” IEEE Transactions on Medical Imaging, Vol. 22, Issue 2, February 2003, pp.189-199.