clear all
fprintf('Reading data...')
data=imread('cameraman.tif');
data=uint8(data);%读入数据,并将数据限制为uint8
fprintf('Done!\n')
%编码压缩
fprintf('compressing data...');
[zipped,info]=norm2huff(data);
fprintf('Done!\n')
%解压缩
fprintf('compressing data...');
unzipped=huff2norm(zipped,info);
fprintf('Done!\n')
%测试是否无失真
isOK=isequal(data(:),unzipped(:))
%显示压缩效果
whos data zipped unzipped
function [zipped,info]=norm2huff(vector)
if~isa(vector,'uint8'),
error('input argument must be a uint8 vector')
end
vector=vector(:)';
%将输入向量转换为行向量
f=frequency(vector);
%计算个元素出现的概率
simbols=find(f~=0);
f=f(simbols);%将元素按出现的概率排列
[f,sortindex]=sot(f);
simbols=simbols(sortindex);
%产生码字 generate the codeword as the 52 bits of a double
len=length(simbols);
simbols_index=num2cell(1:len);
codeword_tmp=cell(len,1);
while length(f)>1,
index1=simbols_index{1};
index2=simbols_index{2};
codeword_tmp(index1)=addnode(codeword_tmp(index1),uint8(0));
codeword_tmp(index2)=addnode(codeword_tmp(index2),uint8(1));
f=[sum(f(1:2)) f(3:end)];
simbols_index=[{[index1 index2]} simbols_index(3:end)];
%将数据重新排列,是两个节点的频率尽量与前一个节点的频率想当
resort data in order to have two nodes with lower frequency as
first to
[f,sortindex]=sort(f);
simbols_index=simbols_index(sortindex);
end
%对应相应的元素与码字
codeword=cell(256:1);
codeword(simbols)=codeword_tmp;
%计算总的字符串长度
len=0;
for index=1:length(vector),
len=len+length(codeword{double(vector(index))+1});
end
%产生01序列
string=repmat(uint8(0),1,len);
pointer=1;
for index=1:length(vector),
code=codeword{double(vector(index))+1};
len=length(code);
string(pointer+(0:len-1))=code;
pointer=pointer+len;
end
%如果需要,加零
len=length(string);
pad=8-mod(len,8);
if pad>0,
string=[string uint8(zeros(1,pad))];
end
%保存实际有用的码字
codeword=codeword(simbols);
codelen=zeros(size(codeword));
weights=2.^(0:23);
maxcodelen=0;
for index 1:length(codeword),
len=length(codeword{index});
if len>maxcodelen,
maxcodelen=len;
end
if len>0,
code=sum(weights(codeword{index}==1));
code=bitset(code,len+1);
codeword{index}=code;
codelen(index)=len;
end
end
codeword=[codeword{:}]
%计算压缩后的向量
cols=length(string)/8;
string=reshape(string,8,cols);
weights=2.^(0:7);
zipped=uint8(weights*double(string));
%存储一个稀疏矩阵
huffcodes=sparse(1,1);% init sparse matrix
for index=1:numel(codeword),
huffcodes(codeword(index),1)=simbols(index);
end
%产生信息结构体
info.pad=pad;
info.ratio=cols./length(vector);
info.length=length(vector);
info.maxcodelen=maxcodelen;
function codeword_new=addnode(codeword_old,item)
codeword_new=cell(size(codeword_old));
for index=1:length(codeword_old),
codeword_new{index}=[item codeword_old{index}];
end
function vector=huff2norm(zipped,info)
%HUFF2NORM Huffman 解码器
%HUFF2NORM(X,INFO)根据信息体结构 info 返回向量 zipped 的解码结果%
%矩阵参数以X(:)形式输入
if~isa(zipped,'uint8'),
error('input argument must be a uint8 vector')
end
%产生01序列
len=length(zipped);
string=repmat(uint8(0),1,len.*8);
bitindex=1:8;
for index+1:len,
string(bitindex+8.*(index-1))=uint8(bitget(zipped(index),bitindex));
end
%调整字符串
string=logical(string(:)');% remove 0 padding
len=length(string);
%解码
weights=2.^(0:51);
vector=repmat(uint8(0),1,info,length);
vectorindex=1;
codeindex=1;
code=0;
for index=1:len,
code=bitset(code,codeindex,string(index));]
codeindex=codeindex+1;
byte=decode(bitset(code,codeindex),info);
if byte>0,%
vector(vectorindex)=byte-1;
codeindex=1;
code=0;
vectorindex=vectorindex+1;
end
end
function byte=decode(code,info)
byte=info.huffcodes(code);
function f=frequency(vector)
%FREQUENCY 计算元素出现概率
if~isa(vector,'uint8'),
error('input argument must be a uint8 vector')
end
f=repmat(0,1,256);
%扫描向量
len=length(vector);
for index=0:256,%
f(index+1)=sum(vector==uint8(index));
end
%归一化
f=f./len;