利用可變編碼比對人形辨識系統實現於DSP平台

Abstract

本論文提出了一組快速而且計算量低的即時人形辨識偵測系統，其用途為在影像監視系統中提供人與非人物體之判別，並且可用在提供紅外線夜視攝影的攝影機上。本系統包含前景取得(Foreground)，人形辨識(Human Detection)，軌跡判別(Trajectory Tracking)等。由於本系統以建立在非支援浮點運算之DSP平台上為前提來進行研究，即時處理(Real-Time)的要求極為嚴苛，計算量以及精準度成為了本論文的第一要求。系統的第一部分在於取出特定場景中的移動前景，在這裡我們使用背景相減法(Background Subtraction)來做為取出移動物體的基礎。為了可以適合各種情況及不同的取像設備，我們使用了一個簡單快速的背景相減二值化闕值(Threshold)設定。第二部分提出了一些簡單的軌跡與狀態判別模式，在之後的人形辨識部分上提供一些必要資訊，以及降低誤判(False-Alarm)，誤判情況偵測等等的處理。而系統的第三部分，人形辨識的判斷法，基於在運算量的要求下，我們在許多的判斷方法之中選擇的以外形樣版為基礎的Codebook判斷方法來實現分辨人形與其他物體的不同。而為了解決室內常有的場景內人物下半身被遮蔽現象，我們提出了Deformable Codebook Matching的方法，可以提供半身以下不同比例的人形辨識機制，以完成雖有部分遮蔽仍有其辨識效果的系統。更進一步的延伸其用法到並排重疊的多人辨識機制。

In this thesis, a fast real-time human detection system with low computing power is proposed. The purpose of this system is to provide the human detection and tracking for video surveillance which can be used in the environment with infrared rays lighting. This system consists foreground segmentation, human tracking, and human detection. The system will be implemented on the real-time DSP system which is not supporting the floating-point computation. The requirement of low computing power and accuracy becomes the major condition that we are very concerned. The first part of our human detection system is to segment the moving object from the scenes. We use the background subtraction here to segment the moving blob. We provide a simple and fast function to calculate the binarization threshold for the varying environments and videos taken by different cameras. In second part of our system, we use simple trajectory tracking and condition judgment to provide some data for human detection algorithm and to decrease the false-alarm rate. The final part is human detection. Because of the requirement of low computing power, we choose the shape-based method by codebook to classify human being from the other objects. The people walking indoor are sometimes covered by furniture such as desks or chairs. To solve this kind of problem, we provide deformable codebook matching, a human detection algorithm for first half body with different height/width ratio. With deformable codebook matching, when someone’s bottom half body is covered, the system can still work. Further, we use deformable codebook matching to implement the human detection for multiple people walking side by side.