CAJ | 학술논문

图片的处理方法多种多样，可以用特征值分解的方法对图片进行信息提取，将此方法与信噪比峰值法、全局质量指标计算法，平均结构化相似度指标的测量法结果作比较，进而证明运用图像特征值分解法的优势。再通过非线性过滤的方法对图片进行进一步的除杂；还可以观察图像频谱中的噪声分布情况，根据这些噪声点的在频率域的分布情况，进而选用合适的滤波器来过滤掉图片中的噪声；为了获得更为清晰的图片，还可以建立模板匹配模型，即：经过有限次的待搜索图与色标模板图的配准过程来完成对目标图片信息的检验和建构，这是此方法较其他方法的优势，也是这篇论文所提倡使用的方法。
도편적처리방법다충다양，가이용특정치분해적방법대도편진행신식제취，장차방법여신조비봉치법、전국질량지표계산법，평균결구화상사도지표적측량법결과작비교，진이증명운용도상특정치분해법적우세。재통과비선성과려적방법대도편진행진일보적제잡；환가이관찰도상빈보중적조성분포정황，근거저사조성점적재빈솔역적분포정황，진이선용합괄적려파기래과려도도편중적조성；위료획득경위청석적도편，환가이건립모판필배모형，즉：경과유한차적대수색도여색표모판도적배준과정래완성대목표도편신식적검험화건구，저시차방법교기타방법적우세，야시저편논문소제창사용적방법。
There are many different ways for the image processing. We can use the method of the eigenvalue decomposition to extract the information of the images. We can compare it with the peak signal-to-noise ratio method, universe quality index method and mean structure similarity index measurement method, hence, proving the advantages of using the eigenvalue decomposition method. Then by the method of nonlinear clarification, we can purify the picture further. We can also observe the distribution of the noise in the frequency spectrum. According to the distribution of the noise points in the frequency spectrum, we can choose a proper filter to filter out the noise in the images. In order to get clearer pictures, we also can construct the matching model of the template. That is, we can test and construct the information of the target images with the limited procedures of matching between the search graph and the color code template figure. It is one of the advantages compared with other methods, also the method this paper recommends.