基于JavaBeans组件技术的工程扫描图像信息处理软件的/The Design and Implementation of an Engineering Sc

2018-08-08 20:44:02

图像 image processing 扫描 scanned



扫描仪是一种常见的图像采集设备,应用领域十分广泛,在不同领域中的应用对扫描图像质量及扫描仪性能要求也不尽相同。为了满足某些专业领域对扫描图像的高质量需求,提高扫描仪产品的性价比,利用数字图像处理技术和专业图像处理软件对扫描图像信息进行后期特殊处理变得越来越重要。
本文对工程扫描图像信息处理中的几个关键技术进行了研究,并根据研究对象的要求采用基于JavaBeans组件技术的框架模型对系统进行了需求分析、总体框架和各功能模块的设计;描述了框架、公共库和模块集之间的关系,给出了公共库为框架和各个模块提供公共属性和表示方法的具体方案及每一个模块封装一组专业图像检测处理算法的策略。本文实现了扫描图像的显示、几何变换、灰度转换、伪色彩增强、亮度/对比度调整以及空间频率域滤波处理等。根据工程扫描图像的大尺寸特性,本文特别研究了对海量图像数据的处理方法,提出了一种数据缓冲区策略,该方法不仅满足了系统处理任意大小扫描图像的需求,而且优化了系统加载海量图像数据的性能,这一点成为该软件系统区别于其他图像处理软件的一个重要标志。此外,本文以简单友好的图形用户界面将图像检测处理结果以数据、曲线及图表等形式输出,方便了用户使用。
本文对所实现的系统进行了综合测试和性能分析,测试结果表明,系统能够满足专业领域扫描图像的处理需求,具有一定的实用价值和应用前景。


Scanner is popular image acquisition equipment widely used in many different fields. The demands for scanned image quality and scanner performance in different areas are respectively diverse. In order to fulfill high quality requirements of scanned images in some special domains and enhance the quality cost ratio of scanner, it is increasingly significant to post-process distinguishingly the scanned image information combining digital image processing techniques and specialized image processing software.

This thesis emphasizes on some key image processing techniques of engineering scanned image and analyses the system demands and designs the system framework and functional modules based on JavaBeans component technology according to the demands of target object. The thesis depicts the relation among framework, common library and module sets and details the scheme by which common library provides public properties and methods for framework and all modules. Each module is encapsulated with a set of professional image detecting and processing algorithms. Some algorithms has been implemented, such as scanned image display, geometric transformation, gray scale transformation, pseudo-color enhancement, brightness/contrast adjustment and special requency domain filter algorithms, etc. According to the large-dimensioned scanned images frequently appeared in engineering, the thesis especially investigates the massive image data processing algorithm and proposes a data buffer strategy. This approach not only meets the requirements of processing arbitrary dimension scanned image but also optimizes the performance of loading massive image data. This has been a distinct characteristic that discriminate this software system from other image processing software. Besides, for the convenience of users, the thesis has developed friendly and simply graphical user interface to output the image detecting result with data, plots and charts.

The thesis tests comprehensively and analyses the performances of the system. The result shows that this system is competent for scanned image processing in target field. The system bears certain practicability and application prospect.