台灣位於環太平洋地震帶上，地形為山地居多，且地質脆弱，加上位於西太平洋副熱帶地區，使得山區常受到颱風的侵擾而發生崩塌，導致土石流和洪水等災害發生，進而影響人民的生命和財產安全。因此，如何有效地建置崩塌地區域資料庫，成為國土保育與災害防治的重要課題。以往利用遙測與航測技術於崩塌地萃取的研究中，大多是於幾何糾正後衛星影像或是航測正射影像上分析崩塌地，但產製正射影像或是糾正衛星影像時，都需要花費較多的時間，對於講求時效性的救災行動而言頗為不利。本研究之目的為發展一套不需使用正射影像萃取崩塌地的方法，以物件導向影像分類法，於DMC（Digital Mapping Camera）航測原始影像上直接萃取崩塌地資訊。首先採取多重解析影像分割的技術，將航測影像依像元光譜和形狀同質性分割成不同區塊（物件），接著利用影像光譜統計值搭配區域成長法，偵測影像中的雲覆蓋地區並過濾。其次，根據光譜亮度統計特徵值，將影像區分成陰暗地區、正常地區以及較亮地區之三種土地覆蓋類型，使用線性相關糾正法（Linear-correlation Correction）將陰暗地區光譜亮度值轉換至正常地區，並利用物件的特徵值，如光譜、面積、形狀以及相關性依序萃取此三種土地覆蓋類型內的崩塌地。最後，使用光線追蹤法 （Ray-tracing），將崩塌地區塊從影像坐標轉換至地圖坐標，使其可以套疊地形資料如坡度、坡向，並進行空間分析以提升崩塌地的判釋精度。研究結果顯示，崩塌地萃取之使用者精度和生產者精度，均有82%以上，並且整個實驗可大量批次處理影像，及快速建立崩塌地資料庫，本研究之方法和崩塌地資料庫將有助於國土保育與崩塌地的災害防治。 / Being located in a subtropical and seismic zone of the West Pacific, the geology is fragile and topography is mountainous in Taiwan. Landslides, floods and other disasters induced by typhoons occur frequently, and it cause the life-threatening and property loss of human beings in Taiwan. Therefore, how to establish landslides data effectively become an important issue of land conservation and disasters management. In recently years, most of the researchers used aerial ortho-images or satellite georeferencing images to detect landslides sites. However, it spent a lot of time generating aerial ortho-images and rectifying satellite images, and it also reduced the efficiency of landslides analysis. Thus, this study developed an object-oriented classification method, which can be directly applied in raw image data, to detect landslides sites. Firstly, this study used multi-resolution image segmentation technique to segment images acquired by Z/I DMC（Digital Mapping Camera） into individual regions （objects） according to the homogeneity of spectral and shape features, and then removed cloud areas by using brightness features depended on the spectral information of images. Secondly, the study divided the entire image into three areas, which are darker area, normal area and lighter area, according to brightness value. Next, Linear-correlation correction （LCC） method was used in this study to transform darker area to normal area so that it can easily detect the landslides sites in darker area, and the object features, such as spectral, area, shape and space correlation indices, were used to extract landslide sites in images. Finally, in order to enhance the accuracy of landslide, the initial landslides were converted from image coordinate system to map coordinate system by ray-tracing method, so the initial landslides data can be further extracted by using topographic data, including slope and aspect data. The results of this study showed that the user and producer accuracies of detecting landslides can reach up to 82%. Moreover, the entire experiments process of this study can batch analyze automatically and establish landslides database quickly. It is expected that the method and landslides data of this study may have contribution to land conservation and disasters management.

本篇論文建立一個藉由直覺性素描搜尋模型的系統，結合筆觸繪圖搜尋手繪圖與模型。希望可以藉由本系統，提供使用者比起關鍵字或模型搜尋模型，更加方便的模型搜尋工具。系統主要分為建立索引檔、比對特徵向量和使用者介面三個部分。建立索引檔部分要將三維模型處理成資料庫可認知的資料型態，首先將模型旋轉到不同角度並且將之從三維空間描繪成二維模型投影圖，再透過分類演算法把模型投影圖和手繪圖描述為二維特徵向量。比對特徵向量部分需建立手繪圖資料庫和三維模型資料庫的橋梁，藉由計算兩者的特徵向量之間的距離與角度，得到相似度的排序。使用者介面部分提供直覺性使用者繪畫的介面，以不影響使用者創造性的前提下，在使用者繪畫過程中給予最相似於使用者繪畫的手繪圖結果，使用者可以藉由臨摹此結果更貼近所想繪畫的物體，更進一步地取得模型的搜尋結果。最後我們將透過統計方法去驗證系統的有效性。 / We proposed an intuitive model retrieval system with a sketch interface for a database contains sketch drawings and 3D models. Benefit the sketch interface, the proposed system can facilitate the search process better than keyword query or search by 3D model. The system begins with offline indexing preprocess which convert the 3D models into feature vectors. Under best view selection, we render each 3D model into a 2D feature line image. Then classification method will apply the line images and sketching images in model database to build the feature vector. The rank of matching is computed with the angle between the feature vector of input sketch image and feature line images in the database. To extend the usability, we design a sketch interface for searching the best match result during the drawing process. For suggesting the drawing hint, candidate matching results are listed aside to the sketch input screen. We use statistical method to evaluate the feasibility of the proposed system.

順應智慧型手機及隨身智慧裝置的普及，人人天天滑手機、對美食拍照、分享社群網站，成為儀式般的全民運動，亦是時下最夯的社會景象。本研究主要描繪影像食物戀者，透過拍攝美食上傳分享的行為，在社群媒介下演進的樣貌。 本研究將依據Meyrowitz媒介情境論及布迪厄資本理論，並透過文本分析法，直接以研究對象過去的美食分享文作分析內容，再以深度訪談法結果為依據，並與文本分析的社群內容作比對，瞭解影像食物戀者的行為演進現象觀察。最後以民族誌參與觀察法，實際驗證研究對象在餐廳用餐時的行為動作記錄，並與深度訪談結果作深度觀察交叉分析。 研究發現，影像食物戀者拍照分享的行為可分為兩大階段、六個歷程，包括基礎演進階段的記錄、分享、學習；以及進階階段的炫耀分享、美食意見領袖及形成自媒體。透過社群媒介的展演，影像食物戀者藉由美食圖文故事的包裝，持續依據自身品味資本在社群上傳播，達到提升個人獨特性及品味層次的目的。 關鍵詞：影像食物戀、媒介情境論、自媒體、社群傳播 / Nowadays, corresponding to the widespread of smartphones and mobile devices, the most popular scenes are that people keep on sliding and checking their mobile phones, taking pictures of fine food, and sharing them in social media communities, which has become a viral phenomenon. This research is based on the Medium Theory by Joshua Meyrowitz and Social Capital Theory by Pierre Bourdieu, in order to understand the behavior evolution of “food shooting addicts,” persons who love to take fine food pictures and upload them to social media communities. Its methodologies include content analysis, in-depth interview, and ethnographic participation and observation. The study finds that there are two stages and six phases in the behavior evolution of food shooting addicts. Primary stage includes: record, share and learn. Advanced stage includes: show-off, becoming opinion leaders and “we media” of fine food. Through their presentation in social media, packaged by beautiful fine food pictures and related stories, food shooting addicts build up their taste capital and achieve the goal of raising up their uniqueness and taste. Keywords: Food Shooting Addicts, Medium Theory, We Media, Social Media Communication

近年來，很多要求高安全性的網站都使用扭曲變形的英文或數字字串作為辨識碼，以避免網站或系統受到大量暴力的攻擊。微軟公司則於2007年提出以貓狗影像的新辨識碼系統—Asirra。對於電腦而言，貓狗影像辨識較字串更為困難。本研究主要針對Asirra的影像資料試圖建構出貓狗影像自動辨識法，藉此來了解此辨識碼系統的有效性。已知影像包含大量雜訊，若使用原始資料則計算困難而且辨識效果差，所以萃取關鍵特徵為重要的研究課題。本文考慮方向梯度直方圖法 (Histograms of Oriented Gradients, HOG) 以及主成分分析 (Principal Components Analysis, PCA) 來篩選重要變數。我們將運用挑選出的特徵建立支持向量機 (Support Vector Machine, SVM) 分類器。在實證分析中，我們發現結合此兩種特徵萃取法，除了能夠大幅降低運算時間，也能得到良好的預測正確率。 / In recent years, many websites, which requires a high standard of security, use CAPTCHA to avoid mass and brutal attacks from hackers. The CAPTCHA considers the use of strings of twisted and deformed English letters or numbers as an identification code. In 2007, the company Microsoft proposed a new image-based recognition system-Assira, which uses dogs and cats images as an identification code. Dogs and cats image recognition is not more difficult than strings of letters or numbers recognition for human, but is more challenging for computers. In this paper, we aim to develop a classification method for images from Asirra. An image is represented by an enormous number of pixels. Only few pixels carry important feature information, most pixels are noise. The abundance of noise leads to computational inefficiency, and even worse, may results in inaccurate recognition. Therefore, in this problem feature extraction is an essential step before a classifier construction. We consider HOG (Histograms of Oriented Gradients) and PCA (Principal Components Analysis) to select important features, and use the features to construct a SVM (Support Vector Machine) classifier. In the real example, we find that combining the two feature detection methods can dramatically reduce computational time and have satisfactory predictive accuracy.

本篇論文意在分析探討葛拉斯的小說《錫鼓》中主角侏儒奧斯卡照片影像式的敘事觀點，以其顛覆且批判德國二次大戰納粹時期的歷史。透過侏儒奧斯卡以回憶述說故事的方式，從被主流歷史掩埋的廢墟殘跡中，提出斷簡殘篇但相互關聯如星宿圖般的照片影像，以存留於照片影像中破碎的個人歷史抵抗正統歷史泯滅真相的力量。由奧斯卡小人物般卑下的目光如攝影鏡頭或伸縮、或放大、或停格的控訴納粹法西斯的惡行，及支持此一罪行的共犯結構— 德國中產階級的狂熱盲從，指摘納粹歷史如場荒誕鬧劇，德國小市民的亂倫敗德與搖旗吶喊的愚昧瘋狂。本論文欲透過班雅明(Walter Benjamin)、蘇珊•宋塔(Susan Sontag)、巴特(Roland Barthes)及德勒茲(Gilles Deleuze)與瓜達里(Félix Guattari)等人之理論架構逐章論述，分析奧斯卡之敘事觀點。首先以其畸形矮小身形猶如攝影式觀看，維持自身與外在世界的批判距離，且將事件做極度怪奇的放大或縮小，試圖還原歷史洪流中被淹沒或為人輕忽的細節。接著透過奧斯卡親人的死亡和他們留存在家庭相簿中的身影，以之視為凝結停頓時空的休止(caesura)和刺點(punctum)，交互出生者與死者，過去與現在，個人與歷史之間的辯證對話。最後探討照片式影像如部份物體(partial object)具有流動、多元、破碎的顛覆特質，檢視奧斯卡的觀點如何逃脫歷史、民族、國家所匯聚的統一整體，解除支配社會場域的「父–母–我」的伊底帕斯關係之力量，揭露納粹如父如神之形象背後的壓抑與荒謬，使其造成的歷史浩劫定格顯影。 / This thesis investigates Gűnter Grass’s The Tin Drum from the perspective of photographic images, using theories drawn from Walter Benjamin, Susan Sontag, Roland Barthes, Gilles Deleuze and Felix Guattari. Acclaimed as the greatest German novel written since the end of World War II, The Tin Drum is supposedly an autobiographical fiction of Oskar Matzerath, a character who, willingly stunting his growth at the age of three, has lived through the long Nazi nightmare and the loss of Danzig City. Wielding his tin drum and piercing scream as weapons, Oskar furnishes the reader with a profound yet hilarious recollection in the form of fragmented yet correlative photographic images on both German history and the human predicament in the modern world. Chapter One explores the effect of alienation resulting from Oskar’s grotesque perspective from Sontag’s idea of photographic seeing and Benjamin’s “distraction” element of the technical media. Oskar observes the adult world, from which he chooses to withdraw, from an unusual angle gained by his dwarfish size. His perspective, likened to the enlargement or close-up of a camera, displays the grotesquery in ordinary events to reveal what is hidden from us. Chapter Two prospects the force of interruption or arrest grounded on Oskar’s bereavement, with Benjamin’s caesura and Barthes’ punctum of photographic images. Death in the photographic images is endowed with the capacity to arrest and petrify the historical movement. The break from the present moment enables the rereading and refiguring of history. The Deleuzo-Guattarian conception of partial objects is employed in Chapter Three to explore how Oskar viewing the external world as objectivized images helps engender revolutionary forces to subvert the orthodox history. The photographic images in the appearance of partial objects break through the totalizing power of orthodox history and bring forth a dialectical optic to examine the past, present and future.

在不利的環境下做交通標誌的偵測與辨識是一件非常艱困的工作，無論在郊區或市區，複雜的環境、天候、陰影以及任何和光線有關的因素甚至是交通標誌遭到遮蔽都將使得偵測與辨識交通標誌變得相當困難。在本篇論文中，我們定義出較寬鬆的顏色分類(color thresholding)方法，配合一些交通標誌的特徵(如外形)來實作出召回率(Recall)較高的偵測系統，另外在辨識方面，最重要的是找出好的辨識特徵，因此我們利用離散餘弦轉換(discrete cosine transform)和奇異值分解(singular value decomposition)處理待辨識標誌擷取其特徵，並配合一些其他的交通標誌特徵，當作類神經網路(ANN)、naïve Bayes classifier等辨識方法的輸入，來幫助我們完成辨識的工作。目前實作出來的系統在有挑戰性的測試資料下有七成六左右的辨識率。 / Robust traffic sign recognition can be a difficult task if we aim at detecting and recognizing traffic signs in images captured under unfavorable environments. Complex background, weather, shadow, and other illumination-related problems may make it difficult to detect and recognize signs in the rural as well as the urban areas. In this thesis, I define a formula for color classification and apply other related features such as the shape of the traffic signs to implement the detection component that offers high recall rate. In traffic sign recognition, the most important thing is to get the effective features. I use discrete cosine transform and singular value decomposition to collect the invariant features of traffic signs that will not be severely interfered by disturbing environments. These invariant features can be used as the input to artificial neural networks or naïve Bayes models to achieve the recognition task. This system yields satisfactory performance about 76% recognition rate when I test them with very challenging data.

數位內容產業即將成為台灣未來下一個重要的策略性產業，對於現有業者與企盼進入的業者，在市場需求快速成長的吸引下，本研究主要目的為利用個案公司之營運策略與規劃，提出相關建議以作為有興趣業者的參考，同時俾利於後續研究者進一步的深入探討。 本研究的主要架構係根據David A. Aaker的理論架構基礎來進行內在與外在分析，同時探討研擬策略規劃方向。整個研究架構分為四個階段：一、探討相關的文獻，以作為研究的原則基礎。二、進行內、外在分析，以界定外在環境中可能發生的機會與威脅，並界定該產業的關鍵成功因素，接著分析個案公司所擁有的優劣勢。三、綜合根據前述之分析，歸納出個案公司所面臨的問題。四、提出可行的策略方向與目標，並擬定個案公司之競爭策略與營運策略。 研究分析結果呈現國內數位內容產業之關鍵成功因素為:擴大內需市場規模、提升產品技術水平、培訓高階專業人才與提供租稅獎勵。而數位內容產業業者之關鍵成功因素為：能夠提供顧客未滿足的需求、將新科技所帶來不同的應用方式延伸、徹底瞭解顧客行業的特性，並從其角度思考、長期上、中、下游的配合關係與信任度、專案管理的執行能力與品質、專案業務累積的經驗與良好口碑、對於長期經營的態度與信念必須誠信及落實、危機趨勢及風險的處理暨因應能力與健全的董事會功能。

本論文應用攝影構圖法則，以模糊邏輯理論為基礎，判別影像的攝影構圖類型。構圖(Composition)乃是攝影這項平面藝術創作最重要的美學元素之一，其目的是利用空間中的物體配置，經由透視投影後，讓畫面的整體呈現平衡感；專業的、優秀的攝影作品，皆會符合攝影的基本構圖原理。因此許多的影像增強、影像合成的應用中，也應該配合相片原本的構圖設計，針對所欲表達的重點予與適當地調整，而非「盲目的」以一體適用的法則去處理每一張照片。 論文中，我們針對影像所欲表達的重點區域，分析其結構特性，設計不同的特徵，並以模糊邏輯理論為基礎，應用Mamdani系統，結合隸屬函數與攝影構圖判別法則的交互作用，用以辨認所欲處理相片的構圖類別。依據辨認後的構圖類別，即可對該影像做適當地分割及調整，以使相片能有最佳的影像增強處理。 實驗證明，本文所提出的方法能有效地辨認攝影構圖類別，針對不同攝影構圖所作的影像修正，才能更符合人眼的視覺喜好。 / This thesis addresses the problem of how to recognize the photographic composition from a given photo based on the theory of fuzzy logic. Composition is one of the important aesthetics for the plane figure photo art. To present the balance of its holistic picture, it takes the advantage of special object arrangement after acting perspective projection. A piece of professional and qualified photo work will realize these basic photo composition methods. For many applications about the digital photo, the operations, i.e., photo enhancement, segmentation, output, and synthesis, all need to match up the photographic composition to do accurate processing rather than “blind” processing that assumes each photo with the same “composition.” An automatic recognition method using image features from some specific regions is described. The method is employed in a Mamdani model and combines outputs of multiple fuzzy logic rules and feature extraction algorithms to obtain confidences that can identify the correct photographic composition. Experimental results show that the proposed method is robust and effective for photographic composition recognition. The feature with adjusting in different photo composing will be able to comfort our human sight.

本研究運用物空間匹配的理論，由已知方位的航空影像與由航空影像自動匹配產生或由空載光達測量取得的數值表面模型(DSM)資料中的平面特徵，製作成正射影像區塊，作為航空影像的控制資訊，並解算待求解之航空影像外方位參數。物空間匹配是以地元(groundel)的概念為基礎，地元為物空間上的單元，每個地元均含描述地表灰值變化的輻射資訊，以及描述地表起伏的幾何資訊。在此一地元即對應於正射影像區塊中的一個正射影像位置，由地元出發，經共線條件並進行輻射資訊調整可於航空影像上產生相對應的像元灰值。依據此關係，產生正射影像區塊時，使用已知方位的航空影像，由DSM資料並萃取物空間中的平面資訊，並由萃取平面上的每一地元出發，經由共線條件比對各已知方位航空影像中相對應像元間的灰值差異，進而產生正射影像區塊；而以正射影像區塊進行影像控制時，則由正射影像區塊中的每一地元出發，仍經由共線條件比對地表上的地元與相對應於待求解之航空影像上的像元灰值差異，達到影像外方位求解之目的。本研究中除以物空間匹配模式發展正射影像區塊製作與航空影像外方位求解之理論之外，並以不同時期的航空影像資料進行實驗與分析。 / Existing aerial images and related spatial data are suitable for control data in aerial photogrammetry. This study will employ the concept of object space matching to generate the orthoimage control patches of plane features from existing aerial images data and digital surface model acquired from aerial image matching or airborne LIDAR surveying. Afterwards, those orthoimage control patches will be used to determine exterior orientation of aerial images by the concept of object space matching. Essentially, object space matching is based on the concept of groundel. The terrain surface condition is described by radiative information and geometric information in each groundel, i.e. each location (or pixel) in orthoimage control patches. The corresponding grey value of each pixel in each aerial image is transformed by the collinear condition and the adjustment of radiative information of groundel. According to the above-mentioned relationship, the relevant theory for generation of orthoimage control patches and exterior orientation determination of aerial images will be developed based on the comparison of grey value between each groundel and each corresponding aerial image pixel. To verify the feasibility of our derived theory, the aerial image data acquired from different period will be utilized for the test. Meanwhile, the relevant problems will be discussed and investigated in the experiments.

本專題為編導式攝影製作，以夢遊為影像故事腳本，隱喻創作由發想至產出的過程。專題概念始於「創作的發想就像一場白日夢，創作的過程就像夢遊」，拍攝六段城市裡的夢遊冒險，將夢遊比喻為創作的不同階段，並在書面文字中解釋這些隱喻。 六段拍攝劇情為「失眠出門、夜遊尋找、甦醒伸展、逃離丟棄、瘋狂吞食，與沈浸熟睡」，由五位主角與我接力演出，以六人表示身處不同創作階段、激發不同面向的自己。故事由主角夜裡失眠開始，進入尋找與丟棄的主線，像是創作初期開啟不同的領域，稍做伸展停留，卻又離棄原有構想，接著進入大吃劇情，以食物帶給身心的雙重滿足，代表創作的喜悅。最後將落葉倒入浴缸中代表歸納與沈浸，浴缸如想像力的船，落葉如眾多故事的隱喻。當我墜入夢境，故事便進入夢中夢的循環。 這是一場呼應自己創作歷程的演出與拍攝，場景與道具皆有隱喻與前後牽引，具影像敘事的實驗性。考量人在夢中的視角，是能夠看見自己演出甚至指揮劇情的，同時具備導演與演員身份，故在作品說明上會使用到第一人稱與第三人稱。 / This project is a production of fabricated photography, which adopts sleepwalk as the script, implying the process of fulfilling thoughts into the tangible work. The project began with the concept that inspiration of creating is similar to daydreaming as well as the journey of creating is akin to sleep-walking. Six sequences of sleepwalk-adventures that shot in the city symbolize the different phases of creating, which will be clarified in the following articles. The plots consist of the insomniac going out, wandering and seeking, waking up and stretching, fleeing and abandoning, wildly swallowing and consuming, and settling and falling asleep, all of which are performed by five characters and me. Six characters indicate that the creator was located and inspired by different phases and facets of creating. The story begins with the protagonist couldn’t get to sleep at night, then wanders into the story line where she/he is seeking and stretching various ideas, then leaving these possibilities behind, which symbolize the first stage of creating. The plots lead us into eating section, in which foods bring both mentally and physically satisfactions to protagonist, imply the gratifications in creating. In the end, pouring the fallen leaves into the tub represents summing up and immersing. The tub is like the boat of imagination carrying fallen leaves that imply diverse stories. Once I fall into the dream land, the story goes into the circle of dreams and dreams. This is a performing and shooting that respond to me. All the sets and properties attaching to one another in the context serve as the holders of metaphors, indicating the experimental essence of image narrative. In terms of viewing angles in the dream, one can see herself/himself in the dream and even dominate the plot of it, as if she/he is playing director and actor at the same time, which is the reason that the first person and the third person narratives are both used in the depictions of this work. Keywords: Creating, Dream to Dream, Fabricated Photography, Image narrative. Sleepwalk.