图像

It has an applicable change of the primitive image matrices, so that the dimension of the image total scatter matrix is further minified. Then, image principal component analysis is used to do image feature extraction.

该方法首先对图像矩阵进行适当的变换，用得到的新的图像矩阵构造图像总体散布矩阵后，再运用图像投影主分量分析进行特征抽取。

Results: The angle that adjusting the reference line on the transverse plane is 32. 85°±6.76°, on the coronal plane is 14.15°±9.02°. The facial canal appears "ㄟ" on double oblique imaging geniculate ganglion, tympanic segment, post lap segment, mastoid segment were showed clearly. The achievement ratio is 100%. The segment displaying on double oblique imaging is more than transverse plane, coronal plane and sagittal plane.

结果：面神经管双斜位图像需要在横断位参考图像上旋转参考线的角度为32.85°±6.76°，在冠状位参考图像上旋转参考线的角度为14.15°±9.02°；在面神经管双斜位图像上，面神经管表现为&ㄟ&形，由前向后同时显示出膝状神经节、鼓室段、后膝部和乳突段；面神经管双斜位成像成功率为100%，面神经管行程的显示程度大于横断位、冠状位和矢状位。

Four images on the image surface respectively correspond space scene information on the four direction and are subjected to character matching and image mosaicking based on the overlap area between the two adjacent scene images, in order to obtain a intact space panoramic image.

像面上四幅图像分别对应空间四个方向上的场景信息，依据相邻场景图像之间的重叠区域进行特征匹配、图像拼接处理，就可以得到一幅完整的空间全景图像。

Noising image is compressed and denoised by integrating Wavelet coding with wavelet threshold denoising together.

将小波图像压缩和小波阈值降噪结合在一起，对噪声图像进行压缩，使得在对图像压缩的同时，实现对图像的降噪。

In order to enhance the image with better anti-noise ability,this paper proposes a image pre-processing algorithm based on lateral inhibition and adaptive filter,which can not only outstands the image'outline,but also eliminates the noise by adaptively changing the model parameters accordingly to the local image's mean,variance and the noise variance.

为了既能增强图像，又具有较好抗噪性能，提出一种基于侧抑制网络和自适应滤波相结合的图像预处理算法，不仅突出了图像的轮廓，而且能根据图像的局部均值和方差以及噪声方差的不同，自适应地改变模型的参数来消除噪声。

He subverts those various symbolic images he intentionally selects in order to verify the detached relationship between image and reality that image is not the one of reality, and reality is not the one in image.

一是他有意识地选择各类象征性图片对之进行颠覆，意在确证图像与现实的梳离关系，图像并非现实的图像，而现实也非图像中的现实。

This dissertation brings forward and realizes one method of establishment of handwritten Chinese character library on account of feature of characters' shape: first, scan the characters wrote on paper to the image file, then break up the whole piece of characters image to solo character, and do some feature extraction. After this, pick up the feature of the image and write in file supervened.

本文提出并实现了基于字形特征的手写体汉字字库的构建方法：先将在纸上书写出的字符扫描成图像文件，然后对整幅的字符图像进行分割，将其分割成一个个的字符，再对图像进行一系列的预处理，接着抽取图像特征，写入文件。

Segmentation is a crucial step transiting from image analysis, and hold important position in image project.

图像分割是由图像过渡到图像分析的关键，在图像工程中占据重要的位置。

Image segmentation is a crucial step transiting from image process to image analysis, and holds important position in image project.

图像分割是由图像处理过渡到图像分析的关键步骤，在图像工程中占据重要的位置。

Firstly, every wavelet base is employed to fuse the multi-source image based on local energy rule, and then a set of the primary fusion images are obtained. Next, these primary fusion images are fused to obtain final fusion image within spatial domain based on local varianc's weighted average rule.

该方法首先用每个小波基对源图像进行基于局部能量的融合，得到多幅初级融合图像，然后对初级融合图像在空间域内再进行基于局部方差加权平均融合，得到最终融合图像。

The split between the two groups can hardly be papered over.

这两个团体间的分歧难以掩饰。

This approach not only encourages a greater number of responses, but minimizes the likelihood of stale groupthink.

这种做法不仅鼓励了更多的反应，而且减少跟风的可能性。