chap10_4

3种定位算法（多边：3 边及4边 最小二乘 质心）的主程序(Location algorithm of three kinds of main program)

Modulation and demodulation BLU

这个是tabu 搜索算法， 用matlab实现，有助于学习tabu算法的人(it is the TABU algorithm)

matlab环境下实现对手写体数字的分类识别(Handwritten Arabia digital numeral recognition)

文中的方法是把图像分块，小波分解得到低频分量、高频分量，然后计算每一块的对比度，把图像块划分为清晰块、模糊块，把清晰块和模糊块相邻的区域定义为边界区域，融合时，直接选取清晰块作为融合后的相应块，对于边界区域，在小波分解的基础上采用基于对比度的像素选取的方法进行处理。(Paper, the method is to image segmentation, wavelet decomposition are low frequency, high frequency components, then calculate the contrast of each piece, the image block is divided into clear blocks, fuzzy block, to clear blocks and fuzzy block is defined as the border region adjacent to area, integration time, a clear block directly select the corresponding block as a fusion, for the border region, the wavelet decomposition on the basis of the pixel-based contrast method selected for processing.)

基于 DCT(离散余弦变换 )域的信息隐藏算法是目前应用最广的变换域算法 ,此类算法的最大不足是嵌信息量小。通过分析 DCT域隐藏算法隐藏容量少的原因 ,提出了一种大容量的 DCT域信息隐藏算法。算法首先对载体图像进行频谱均匀化处理 ,然后对图像进行全局 DCT变换 ,并将隐藏信息嵌入到载体图像的低频 DCT系数上 ,信息嵌入采用分段量化的方法实现 ,提取隐藏信息时不需要原始载体图像。实验结果表明 ,该算法的嵌入容量大 ,不可感知性好。(Based on DCT (discrete cosine transform) domain information hiding algorithm is the most widely used transform domain algorithms, the largest of these algorithms is the lack of a small amount of information embedded. Hiding Algorithm for DCT domain by analyzing the hidden reasons for the low capacity, a large capacity of information hiding algorithm for DCT domain. Firstly, the spectrum of the carrier image homogenization, and then the image is a global DCT transformation, and hidden information embedded into the image of the low-frequency DCT coefficients, the information embedded in the methods used to achieve sub-quantization, extract hidden information without the original Vector image. Experimental results show that the algorithm embedding capacity, can not be perceived good.)

此算法为基于信息熵的免疫算法求TSP的源程序!(algorithm based on information entropy of the immune algorithm for TSP source!)

Fuzzy knowledge-based system identification and modeling process is composed of two parts: variable identification and parameter identification. In variable identification, the significant variables of the system are identified among the set of possible variables. In parameter identification, the parameters of the knowledge-based system that describe the relationship between input and the output variables are identified. Parameter identification method consist of two steps: 1-find fuzzy clusters to establish the number of fuzzy rules 2-optimize the rule consequents. This code perform a fuzzy clustering using subtractive clustering

this program will help you to pick a random number out of an array

good refernce for matlab