「構圖」是創造性的發揮自己所想表達的方法，也是畫面的布局與結構的安排。攝影構圖是視覺效果的關鍵因素，透過構圖，攝影師能夠讓影像傳遞出所想表達的資訊，吸引觀眾的目光。雖然構圖沒有絕對的定式和規律，但是客觀的存在著一些形式上的法則，例如三分法則、對稱等等。

有別於先前的研究大多是對影像進行拍攝後的分析，本論文提出並且設計了一款攝影的應用程式，能夠即時的針對使用者選擇的構圖技巧，進行影像畫面的評分，讓使用者能藉由該應用程式，在攝影的過程中，掌握攝影構圖理論的基礎，並了解在攝影創作中，對藝術創造起作用的要素與原理，以及其視覺效果。本論文總共選擇八種常見的構圖技巧進行分析：水平構圖、三分法則、三角構圖、匯聚線、框架式、強度平衡以及明暗對比。

"Composition" is a creative way for expressing what you want through the layout and structure of an image. Photograph composition is a key value for visual effects. Through composition, photographers can let images convey the information that they want to express, and attract the attention of the audience. Although there is no absolute formula and rule in composition, objectively there are some formal rules, such as rule of thirds and symmetry and so on.

Different from previous researches, where most of them perform analysis after the images are taken. In this paper, we propose and design a photography application, which can rate the frame according to the user-selected composition setting. This helps a user know the basis of photography composition theory in the process of photograph, and understand the elements and principles that play important roles in photographic creation, as well as in its visual effect. In this paper, eight common composition criteria are selected for analysis: horizon, rule of thirds, triangle, convergence, frame, intensity balance and contrast.

緒論
文獻探討
演算法設計與系統實作
系統結果展示
系統實驗
結論與未來展望

Personalized Photograph Ranking and Selection Systems
A Computational Approach to Edge Detection
Use of the Hough Transformation To Detect Lines and Curves in Pictures
Mean Shift, Mode Seeking, and Clustering
Topological Structural Analysis of Digitized Binary Images by Border Following
An Optimal Algorithm for Finding Minimal Enclosing Triangles
Road Vanishing-Point Detection: A Multi-Population Genetic Algorithm Based Approach
"GrabCut": interactive foreground extraction using iterated graph cuts
Rapid Object Detection using a Boosted Cascade of Simple Features
https://jacobgil.github.io/computervision/saliency-from-backproj
https://burst.shopify.com/