基于DM6437的圖像平滑算法的設計與實現

電子信息工程論文編號:DZXX332  論文字數:20706,頁數:57
目  錄

中文摘要、關鍵詞……………………………………………………………………………Ⅰ
英文摘要、關鍵詞……………………………………………………………………………Ⅱ
引言……………………………………………………………………………………………1
第一章 圖像平滑 ……………………………………………………………………………2
 1.1 圖像噪聲的基本概念…………………………………………………………………2
 1.2 圖像平滑技術…………………………………………………………………………3
第二章 DSPs芯片 ……………………………………………………………………………9
 2.1 DSPs芯片概述 ………………………………………………………………………9
 2.2 TMS320C6000系列 DSPs……………………………………………………………13
第三章 系統硬件設計……………………………………………………………………16
 3.1 DSP 系統的基本硬件設計…………………………………………………………16
 3.2 DM6437………………………………………………………………………………20
第四章 程序設計與實現…………………………………………………………………23
 4.1 軟件開發環境 CCS…………………………………………………………………23
 4.2 程序設計與實現……………………………………………………………………23
結 論……………………………………………………………………………………………33
致 謝……………………………………………………………………………………………34
參考文獻………………………………………………………………………………………35
附 錄……………………………………………………………………………………………36

基于DM6437的圖像平滑算法的設計與實現

摘要：圖像是人類獲取和交換信息的主要來源，但是在圖像的形成、傳輸和變換過程中，由于受多種因素的影響，往往使圖像與原始景物或圖像與原始圖像之間產生某種差異，使得圖像通常模糊不清，使人觀察起來不滿意，或者使機器從中提取的信息減少甚至造成錯誤。因此，必須對這些圖像進行改善。
 圖像平滑處理是一種圖像增強技術，其目的是減少圖像噪聲。圖像的常見噪聲主要有加性噪聲、乘性噪聲等。圖像中的噪聲往往和信號交織在一起，尤其是乘性噪聲，如果平滑不當，就會使圖像本身的細節如邊界輪廓、線條等變得模糊不清。如何既平滑掉噪聲又盡量保持圖像細節，是圖像平滑研究的主要任務。
 論文中介紹了圖像噪聲產生的原因和噪聲的分類，詳細的闡述了均值濾波、中值濾波、巴特沃斯低通濾波這三種常用圖像平滑技術的原理和方法。接著介紹了DSPs，設計了DSP的基本硬件系統，對DM6437開發平臺做了詳細介紹。然后對三種圖像平滑技術進行了算法設計，并在CCS3.3開發環境下，用DM6437開發平臺實現了圖像平滑。最后從程序執行時間、圖像峰值信噪比兩個方面對三種平滑技術進行了分析與比較。最終得出了這三種平滑技術各自的優缺點。
關鍵詞：圖像平滑  中值濾波  均值濾波  巴特沃斯低通濾波  DM6437

The Design and Implementation of
Image Smoothing Algorithm based on DM6437

Abstract：Image is one of the main sources for mankind to acquire and exchange information. But during the image of formation, transmission and transformation, it often has some differences between the original image and the real natural scene because of many factors. People are usually not satisfied with the image for its blurring. Information extraction from machine is also a way for this blurring. So, we must to find some methods to improve the quality of the image.
 Image smoothing technology is an image strengthening method which aims to reduce noise from the image. The common noises of image mainly contain additive noise, multiplicative noise, etc. The noise in image and signal are always intertwined with each other, especially in multiplicative noise. If not smooth the image properly, it will make the details of the image itself, such as the boundary contour, lines and so on, get blurring . How to smooth out the noise while preserving the detail of the image as much as possible is the main task of image smoothing technology.
 The present Paper describes the causes of image noise and noise classification and discusses the principles and methods of the mean filter, median filter and Bathurst low-pass filter in detail. The paper then introduced the DSPs, designed the basic hardware system of the DSP and discussed the DM6437 development platform in detail. From the program execution time and the image peak of SNR analyzed and compared. Finally we obtained their advantages and disadvantages of two smoothing technology by my analysis and comparison.
 Key words: Image Smoothing; Median-Filter; Mean-Filter; Butterworth LPF; DM6437