本書主要內(nèi)容圍繞3D 計算機視覺展開，介紹了相關(guān)的基礎(chǔ)概念、基本原理、典型算法、實用技術(shù)和應(yīng)用成果。本書可在學過其姊妹篇《2D 計算機視覺：原理、算法及應(yīng)用》后學習。 本書將從客觀場景出發(fā)到后對場景進行理解的全過程分為5 個部分進行介紹。第1 部分是圖像采集，介紹了攝像機標定和3D 圖像采集技術(shù)；第2 部分是視頻運動，介紹了視頻圖像和運動信息，以及對運動目標進行檢測和跟蹤的技術(shù)；第3 部分是物體重建，介紹了雙目立體視覺和單目圖像恢復(fù)技術(shù)；第4 部分是物體分析，介紹了3D 目標表達和廣義匹配；第5 部分是高層理解，介紹了知識和場景解釋及時空行為理解。本書除提供大量示例外，還針對每章的內(nèi)容提供了自我檢測題（含提示并附有答案），并且給出了相關(guān)的參考文獻和術(shù)語索引（包括英文）。
圖像處理領(lǐng)域?qū)＜艺仑箷x教授全新力作；
理論講解 應(yīng)用示例；
內(nèi)容全面，講解透徹，全篇干貨。
本書是一本介紹3D 計算機視覺基本原理、典型算法和實用技術(shù)的專業(yè)圖書，建議讀者在學習《2D 計算機視覺：原理、算法及應(yīng)用》后使用本書。
本書在選材上主要覆蓋了計算機視覺的進階級內(nèi)容，自成體系，主要針對信息類相關(guān)專業(yè)，同時兼顧了具有不同專業(yè)背景的學習者及自學讀者的需求。讀者可據(jù)此開展科研工作并解決實際應(yīng)用中一些具有一定難度的問題。
本書在編寫上比較注重實用性，沒有過多強調(diào)理論體系，盡量減少公式推導(dǎo)，著重介紹常用的方法。書中有較多的示例，能通過直觀的解釋幫助讀者理解抽象的概念。書末附有術(shù)語索引（文中標為黑體），給出了對應(yīng)的英文，方便讀者查閱及搜索相關(guān)資料。
本書提供了大量的自我檢測題（包括提示和答案）。從目的來說，一方面，這便于自學者判斷自己是否掌握了重點內(nèi)容；另一方面，這便于教師開展網(wǎng)絡(luò)教學，在授課時加強師生互動。題目類型為選擇題，可用計算機方便地判斷正誤。從內(nèi)容來看，很多題把基本概念換一種說法進行表達，補充了正文，使學習者能加深理解；有些題列出了一些相似但不相同（甚至含義相反）的描述，通過正反辯證思考，使學習者能深入領(lǐng)會本質(zhì)。所有自我檢測題都附有提示，讀者可獲得更多的信息以進一步理解題目的含義。同時，在有提示的基礎(chǔ)上，如果讀者能在看到提示后完成自我檢測題，則表明基本掌握了學習內(nèi)容；如果不看提示就能完成自我檢測題，則表明內(nèi)容掌握得比較好。
本書從結(jié)構(gòu)上看，包括12 章正文、1 個附錄及自我檢測題、自我檢測題答案、參考文獻和術(shù)語索引。在這17 個一級標題下，共有66 個二級標題（節(jié)），再之下有135 個三級標題（小節(jié)）。全書共有文字（包括圖片、繪圖、表格、公式等）50 萬字，共有編了號的圖228 個、表格22 個、公式565 個。為便于教學和理解，本書給出示例68 個、自我檢測題157 道（全部附有提示和答案）。另外，書末列出了直接相關(guān)的100 多篇參考文獻和用于索引的500 多個術(shù)語（中英文對照）。
本書的先修課程知識涉及三個方面。一是數(shù)學，包括線性代數(shù)和矩陣理論，以及有關(guān)統(tǒng)計學、概率論和隨機建模的基礎(chǔ)知識；二是計算機科學，包括對計算機軟件技術(shù)的掌握、對計算機結(jié)構(gòu)體系的理解，以及對計算機編程方法的應(yīng)用；三是電子學，包括電子設(shè)備的特性原理及電路設(shè)計等內(nèi)容。
感謝電子工業(yè)出版社編輯的精心組稿、認真審閱和細心修改。
后，作者感謝妻子何蕓、女兒章荷銘在各方面的理解和支持。
章毓晉
2020 年暑假于書房
章毓晉，于1989年獲比利時列日大學應(yīng)用科學博士學位。19891993年，先后在荷蘭德爾夫特大學從事博士后研究工作并擔任專職研究員。于1993年到清華大學任教，1997年被聘為教授，1998年被聘為博士生導(dǎo)師，2014年成為教學科研系列長聘教授。在2003年學術(shù)休假期間，同時被聘為新加坡南洋理工大學訪問教授。在清華大學，先后開出并講授10多門本科生和研究生課程。在南洋理工大學，開出并講授研究生課程現(xiàn)代圖像分析（英語）。已編寫出版了圖像工程系列教材第1版、第2版、第3版和第4版，《圖像工程問題解析》、《圖像處理和分析基礎(chǔ)》、《圖像處理和分析技術(shù)》（第2版和第3版）、《圖像處理和分析教程》（第1版、第2版和第3版）、《計算機視覺教程》（第1版、第2版和第3版）和《圖像處理基礎(chǔ)教程》，以及Image Engineering: Processing, Analysis, and Understanding，Image Engineering, Vol.1, Image Processing，Image Engineering, Vol.2, Image Analysis，Image Engineering, Vol.3, Image Understanding，翻譯出版了《彩色數(shù)字圖像處理》、《圖像處理基礎(chǔ)》（第2版）、《MATLAB圖像和視頻處理》、《計算機視覺基礎(chǔ)》和《彩色計算機視覺》，研制出版了《圖像處理和分析多媒體計算機輔助教學課件》和《圖像處理和分析網(wǎng)絡(luò)課程》。已在國內(nèi)外發(fā)表了30多篇教學研究論文。主要科學研究領(lǐng)域為所提出的圖像工程（圖像處理、圖像分析、圖像理解及其技術(shù)應(yīng)用）。自1996年起，已連續(xù)26年對中國圖像工程的研究及主要文獻進行了系統(tǒng)的年度分類和總結(jié)綜述。已在國內(nèi)外發(fā)表了500多篇圖像工程研究論文，出版了專著《圖象分割》《基于內(nèi)容的視覺信息檢索》《基于子空間的人臉識別》，編著了《英漢圖像工程辭典》（第1版、第2版和第3版）、《圖像工程技術(shù)選編》和《圖像工程技術(shù)選編（二）》，主持編著了Advances in Image and Video Segmentation，Semantic-Based Visual Information Retrieval，Advances in Face Image Analysis: Techniques and Technologies，出版了Handbook of Image Engineering和A Selection of Image Processing Techniques.曾任第24屆國際圖像處理學術(shù)會議（ICIP2017）等20多個國內(nèi)外學術(shù)會議的程序委員會主席�，F(xiàn)為中國圖象圖形學學會名譽監(jiān)事長和會士、國際電氣電子工程師協(xié)會（IEEE）高級會員、國際光學工程協(xié)會（SPIE）會士（因在圖像工程方面的成就）。
第1 章計算機視覺概述·····································································.1
1.1 人類視覺及特性····································································.1
1.1.1 視覺特點····································································.2
1.1.2 視覺的亮度特性···························································.3
1.1.3 視覺的空間特性···························································.5
1.1.4 視覺的時間特性···························································.6
1.1.5 視知覺·······································································.8
1.2 計算機視覺理論和框架··························································.10
1.2.1 計算機視覺的研究目的、任務(wù)和方法·······························.11
1.2.2 視覺計算理論·····························································.11
1.2.3 框架問題和改進··························································.16
1.3 3D 視覺系統(tǒng)和圖像技術(shù)·························································.18
1.3.1 3D 視覺系統(tǒng)流程·························································.18
1.3.2 計算機視覺和圖像技術(shù)層次···········································.19
1.3.3 圖像技術(shù)類別·····························································.20
1.4 本書結(jié)構(gòu)框架和內(nèi)容概況·······················································.21
1.4.1 結(jié)構(gòu)框架和主要內(nèi)容····················································.22
1.4.2 各章概況···································································.22
1.5 各節(jié)要點和進一步參考··························································.23
第2 章攝像機標定···········································································25
2.1 線性攝像機模型···································································.25
2.1.1 完整成像模型·····························································.26
2.1.2 基本標定程序·····························································.27
2.1.3 內(nèi)、外參數(shù)································································.28
2.2 非線性攝像機模型································································.30
2.2.1 畸變類型···································································.31
2.2.2 標定步驟···································································.33
2.2.3 標定方法分類·····························································.34
2.3 傳統(tǒng)標定方法······································································.35
2.3.1 基本步驟和參數(shù)··························································.36
2.3.2 兩級標定法································································.36
2.3.3 精度提升···································································.40
2.4 自標定方法·········································································.41
2.5 各節(jié)要點和進一步參考··························································.44
第3 章 3D 圖像采集··········································································45
3.1 高維圖像············································································.45
3.2 深度圖···············································································.46
3.2.1 深度圖和灰度圖像·······················································.47
3.2.2 本征圖像和非本征圖像·················································.47
3.2.3 深度成像方式·····························································.49
3.3 直接深度成像······································································.50
3.3.1 飛行時間法································································.50
3.3.2 結(jié)構(gòu)光法···································································.53
3.3.3 莫爾等高條紋法··························································.55
3.3.4 同時采集深度和亮度圖像··············································.58
3.4 立體視覺成像······································································.59
3.4.1 雙目橫向模式·····························································.59
3.4.2 雙目會聚橫向模式·······················································.63
3.4.3 雙目軸向模式·····························································.65
3.5 各節(jié)要點和進一步參考··························································.67
第4 章視頻圖像和運動信息·······························································69
4.1 視頻基礎(chǔ)············································································.70
4.1.1 視頻表達和模型··························································
書單推薦
