基于DSP平台及RF5框架的图像增强系统的实现

发布时间：2018-06-03 07:52

本文选题：参考框架RF5 + 开发平台TMS320DM642　；参考：《内蒙古大学》2017年硕士论文

【摘要】：现如今,数字电子技术发展越来越迅速,数字信号处理器的普及率越来越高,应用范围随之也越来越广泛,DSP数字信号处理器在数字电子领域已经起了至关重要的作用。图像增强,在提高图像质量的图像处理中有着举足轻重的作用,它选择想要的信息着重强调,对于另一些信息可以抑制,因此可以优化图像的观测效果,可以使得增强后的图像对于人的观测分析和计算机的计算处理更加友好。图像增强的方法分为空域增强法和频域增强法两种,在频域增强法中,因小波变换具有良好的局部时间频率特性,所以其在数字图像处理领域中得到了的广泛使用。本文采用 TMS320DM642 DSP 开发平台,RF5(Reference Frameworks 参考框架),实现对图像的加载、增强、显示的算法操作,实现系统结构化开发的规范化。为了解决传统DSP编程难度大和消耗时间长等缺点带来的低效和复杂问题,本文还采用Matlab/Simulink、CCSV3.3通过代码自动生成技术来实现小波变换图像增强算法。与ARM,FPGA等嵌入式工具相比较,DSP由于采用哈佛结构,集成专门的硬件乘法器,能够轻松实现大数据运算,实现快速的实时数据处理。RF5是一种参考框架,是eXpressDSP软件的关键组件,并且利用RF5开发的软件结构在一定范围内可以通用。由于RF5的使用,使得算法和系统软件的接口变的标准化,并且减轻了系统的集成任务,极大程度的缩短了相应的开发周期。论文的最后,通过主观、客观方式分别对通过本设计实现的增强系统处理后的图片进行测评,检验系统和算法的合理性以及实用性,并且对本系统的发展提出了完善建议与未来展望。
[Abstract]:Nowadays, with the rapid development of digital electronic technology, the popularity of digital signal processor (DSP) is becoming higher and higher, and the application of DSP is becoming more and more extensive. DSP has played an important role in the field of digital electronics. Image enhancement plays an important role in image processing that improves image quality. It chooses the desired information to emphasize that other information can be suppressed, so that the observation effect of the image can be optimized. It can make the enhanced image more friendly to human observation and computer processing. Image enhancement methods are divided into spatial domain enhancement and frequency domain enhancement. In frequency domain enhancement, wavelet transform is widely used in digital image processing because of its good local time and frequency characteristics. In this paper, the TMS320DM642 DSP development platform RF5 / reference Frameworks reference frame is used to realize the algorithm operation of image loading, enhancement and display, and the standardization of system structure development is realized. In order to solve the problem of low efficiency and complexity caused by the disadvantages of traditional DSP programming, such as high difficulty and long time consumption, this paper also uses Matlab / Simulink DSP CCSV3.3 to realize wavelet transform image enhancement algorithm by code automatic generation technology. Compared with embedded tools such as ARMN FPGA, DSP is a reference framework and a key component of eXpressDSP software because it adopts Harvard structure and integrates special hardware multiplier, which can easily realize big data operation and realize fast real-time data processing. And the software structure developed with RF5 can be used in a certain range. Because of the use of RF5, the interface between the algorithm and the system software is standardized, and the integration task of the system is reduced, and the corresponding development cycle is greatly shortened. At the end of the thesis, through subjective and objective methods, the images processed by the enhanced system are evaluated, and the rationality and practicability of the system and algorithm are tested. And put forward the perfect suggestion and future prospect to the development of this system.
【学位授予单位】：内蒙古大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP391.41

【参考文献】