图像

Medical image processing enhancing the image knowability, thereby, it is more satisfying to observe and process image. Medical image enhancement mostly selectively extrude the interested characters of image, give prominence to profile of targets, enhance the details of image, advance the arrangements, filtrate all kinds of yawps and so on.

医学图像增强主要是为了将图像中感兴趣的特征有选择的突出，具体表现为突出目标物轮廓，增强图像中的细节，提高层次感，滤除各类噪声等等。

Experiment proved that the means can linearize the camera model and be the foundation to account out the position of target.

为了对机器人场景进行描述，本文提出了基于颜色的图像分割与目标定位技术：根据彩色图像的颜色特征对图像进行分割，运用快速聚类法对分割结果分类。

The initial threshold could be chosen by using Otsu algorithm, and in the iteration process, an optimal threshold was determined by maximizing the MI between the original image and the segmented image.The effectiveness of the proposed approach was evaluated by applying it to the microphotographic.

首先利用Otsu算法确定全局阈值作为初值，以互信息量为目标函数，在小范围内计算分割图像与原图像的互信息量，互信息量达到最大时的阈值即为最优值，这是将图像配准方法用于分割的一种创新性尝试。

This method consists of two steps: first, the image is divided into numbers of nonoverlapping blocks and the GMRF parameters are computed for each of these blocks, then a clustering method is used to merge these blocks, a coarse segmentation and texture types of the image are obtained. Second, a deterministic relaxation neural network is constructed, the parameters of the model estimated from the first step are used as input data, through neural network computation, a fine segmentation is obtained.

提出一种基于高斯-马尔可夫随机场和神经网络的无监督纹理分割方法，方法分为两步：第一步先将图像分为不重叠的小块，在小块中计算GMRF参数，并将此参数和该块的均值、方差作为该块的特征向量，然后进行聚类，得到原图像的一个初始分割和图像中所包含的类别数；第二步构造一个决定性松弛的神经网络，将第一步得到的结果作为初始输入，经过神经网络计算，得到一个精确的分割结果。

Multi-channel linear-phase nonuniform filter banks with proper sampling factors are designed which are applied to rows and columns of the input image, respectively, according to its power spectrum distribution. By this procedure, the image can be segmented nonuniformly in the frequency domain.

根据输入图像的频谱能量分布和实际应用的需要，设计具有合适采样因子的线性相位非均匀滤波器组并将其应用于对图像行、列的处理中，实现对整个图像按照频谱能量的高低进行灵活分割的过程。

The image reconstruction algorithms is thoroughly researched. The Filter Back Projection , Algebra Reconstruction Technology and fan beam data rearrangement algorithm used in medical computer tomography are improved. The NOSER algorithm, linear neural networks method and Hopfield neural networks method are presented and gain the better result of image reconstruction.

深入研究了图像重建算法，改进了医学CT的滤波反投影算法、代数重建技术和扇束投影数据重排方法，提出了带图像光滑约束的NOSER算法、线性神经网络方法和Hopfield神经网络方法，并得到了较好的图像重建结果。

This method was made up of a series of image segmentation and classification technique. First, the orthograph revised SPOT image was segmented by applying a region-based segmentation algorithm. Based this reference map, TM image was classified using the maximum likelihood classifier and additional empirical rules.

该方法由图像分割和分类等一系列技术组成，首先用基于区域分割法则对正射校正SPOT图像进行分割，然后把它作为参考用最大似然法分类器和其他一些经验规则对TM图像进行分类。

This method was made up of a series of image segmentation and classification technique. First, the orthograph revised SPOT image was segmented by applying a regionbased segmentation algorithm. Based this reference map, TM image was classified using the maximum likelihood classifier and additional empirical rules.

该方法由图像分割和分类等一系列技术组成，首先用基于区域分割法则对正射校正SPOT图像进行分割，然后把它作为参考用最大似然法分类器和其他一些经验规则对TM图像进行分类。

A translation-invariant extension of multiwavelet transform called stationary multiwavelet transform was proposed,which yields an overcomplete signal representation,and consequently incorporates the advantages of multiwavelet transform and translation-invariant property.

综合了多小波分析和平移不变性质的优势，将多小波分析扩展到静态多小波的范畴，提出了一种基于静态多小波变换的图像融合方法。通过对源图像进行SMWT，根据变换系数的尺度内多子带联合窗口特性，实现多源图像的融合。

It is used in B-ultrasound medical equipment, aviation helmet system as display terminal in the photic information dealing system.

为医用B超系统设计的。该显示器具有两种图像功能--隔行图像扫描和逐行图像扫描，手动切换。

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。