透過您的圖書館登入
IP:3.145.167.226
  • 期刊
  • OpenAccess

應用影像處理技術輔助植物葉片之幾何模擬

Geometric Modeling of Plant Leaves Aided with Image Processing Techniques

摘要


本研究結合影像處理與計算機繪圖技術,建立了一套植物葉片模型的幾何模擬方法與軟體,此方法可以快速地由葉片的影像擷取葉形,並利用自動配適建立葉片的貝氏曲線模型,在不同的逼近程度下模擬植物葉片的形狀。我們首先是以兩個貝氏曲線來描述葉形,再進一步將平面的模型立體化,重建成貝氏曲面立體葉片模型。應用此種幾何模擬的方法,對十四種典型的葉形進行測試,利用兩條貝氏曲線的合成可對全緣且葉尖無變化之線形、披針形、卵形、橢圓形、長橢圓形、帶形、心形、菱形、圓形、腎形共十種葉形葉片做良好之外型描述。其平均面積誤差在5%以內,平均逼近誤差皆在10%以內,且平均周長誤差亦在4%以內。對於三角形、匙形、齒匙形、提琴形等四種葉形來說,其誤差較大,這些類型的葉片宜以多段的貝氏曲線進行幾何模擬,以準確地逼近實際的葉片形狀。本文中詳細介紹了由植物葉片透過影像處理技術快速建立二維與三維葉片模型的方法,同時亦以植物結構的計算機繪圖模擬為例,說明植物葉片模型在虛擬植物系統之應用。

並列摘要


A geometric modeling method and software based on image processing technique and computer graphics were established to describe various types of plant leaves. The leaf shape can be effectively acquired from leaf image and converted to Bezier curves using automatic fitting to the leaf boundary points. The constructed model can be used to describe leaf shape with various degrees of accuracy. The geometric model was mainly composed of two Bezier curves that can be further developed into Bezier surfaces for 3-dimensional model. Fourteen typical leaf shapes were tested with this modeling approach and the results revealed that ten of them - lanceolate, ovate, elliptical, oblong, cordate, orbicular, linear, fascia, rhombate and reniform - can be described with two Bezier curves with sufficient accuracy. The area error, fitting error and contour error between the real and modeled leaf shapes were less than 5%, 10% and 4%, respectively. For the other four leaf shapes - spathulate, denta-spathulate, cello-shaped and deltoid - more segments of Bezier curves were necessary for a good fitting. This paper meticulously describes the image processing algorithms to extract leaf shape, and the methods in constructing the 2-dimensional and 3-dimensional geometric models. The applications of the leaf shape model were also demonstrated with examples of constructing virtual plants using computer graphics.

被引用紀錄


蔡馨儀(2012)。以地面光達資料重建大葉桃花心木三維模型〔碩士論文,國立屏東科技大學〕。華藝線上圖書館。https://doi.org/10.6346/NPUST.2012.00183
楊格(2012)。利用類神經網路於浮游藻類自動影像辨識分類之研究〔碩士論文,國立臺灣大學〕。華藝線上圖書館。https://doi.org/10.6342/NTU.2012.01551

延伸閱讀