%?function?superbest=star_opso（mnd）?
m=5;?
n=10;?
d=5;?
%基于組織的粒子群優化算法入口?
%m個組織，每個組織中有m個粒子?
%n為最大迭代次數?
%d為支撐基維數?
%函數opso-fun可改變，但同時支撐基的維數d也會改變?
%當函數OPSO_fun改變時注意函數opso_data和函數opso_vect中支撐基維數d的改變?
im=imread（‘lena.bmp‘）;??%讀入待分割圖像?
if（?isrgb（?im?）?）?
????im?=?rgb2gray（?im?）;?
end;?
[immimn]=size（im）;??%計算待分割圖像的大小?
figure（1）;imshow（im）;title（‘Original‘）;?
?
x=opso_data（m）;??%x為粒子，初始化m*m個粒子的位置?
v=opso_vect（m）;???%v為速度，初始化m*m個粒子的速度?
wmax=0.900000000001;?????%權重最大值?
wmin=0.4;?????%權重最小值?
c1=2.05;???%常系數?
c2=2.05;???%常系數?
for?i=1:m?
????for?j=1:m?
????????for?h=1:d?
????????????xx（h）=x（ijh）;?
????????end?
????????x（ijd+1）=opso_fun1（xxdouble（im）immimnd）;??????%計算每個粒子的值，并賦予粒子后?
????????v（ijd+1）=0;?
????end?
end?
for?i=1:m?
????for?j=1:m?
????????for?h=1:d+1?
???????????persongood（1ijh）=x（ijh）;??????%粒子個體最優值初始化?
???????????persongood（2ijh）=v（ijh）;??????%粒子個體最優速度初始化?
????????end?
????end?
end?
allbetter=opso_all_better（persongoodmd）;??????%查找每個組織的最優值?
j=1;????????%查找整個組織最優值（小）?
for?i=1:m?
????if?allbetter（1id+1）<=allbetter（1jd+1）?
????????j=i;?
????end?
end?
y1=allbetter（1jd+1）;???????%初始整個組織最優值（小）賦值?
y2=y1;?
for?k=1:n????????%開始迭代運算，k為迭代次數，n為最大迭代次數?
????w=wmax-（wmax-wmin）*k/n;?
????for?i=1:m?
????????for?j=1:m?
????????????for?h=1:d?
????????????????r1=rand（11）;?
????????????????r2=rand（11）;?
????????????????v（ijh）=w*v（ijh）+c1*r1*（persongood（1ijh）-x（ijh））+c2*r2*（allbetter（1ih）-x（ijh））;????%更新粒子速度?
????????????????x（ijh）=x（ijh）+v（ijh）;?????%更新粒子位置?
????????????end?
????????end?
????end?
????for?i=1:m?
????????for?j=1:m?
????????????for?h=1:d?
????????????????xx（h）=x（ijh）;?
????????????end?
????????????x（ijd+1）=opso_fun1（xxdouble（im）immimnd）;??????%計算更新后每個粒子的值，并賦予粒子后?
????????end?
????end?
????s=0;??????%記錄粒子個體最優值是否更新?
????for?i=1:m?
????????for?j=1:m?
????????????if?x（ijd+1）<=persongood（1ijd+1）?
???????????????for?h=1:d+1?
???????????????????persongood（1ijh）=x（ijh）;??????%更新粒子個體最優值?
???????????????????persongood（2ijh）=v（ijh）;??????%更新粒子個體最優速度??
???????????????end?
???????????????s=s+1;?
????????????end?
????????end?
????end?
????if?s~=0???????%當有粒子個體最優值更新時，計算組織的最優值（小）?
????????allbetter=opso_all_better（persongoodmd）;??????%計算更新每個組織的最優值?
????????t=1;??????%查找迭代更新后組織最優值（小）?
????????for?i=1:m?
????????????if?allbetter（1id+1）<=allbetter（1td+1）??????
????????????????t=i;?
????????????end?
????????end?
????????y2=allbetter（1td+1）;???????%整個組織最優值（小）賦值?
????end?
????if?y1~=y2?????%當整個組織最優值更新時，極值最差的組織遭到淘汰，隨之產生強勢組織?
????????h=1;??????%查找整個組織中最差（大）組織h?
????????for?i=1:m?
????????????if?allbetter（1id+1）>=allbetter（1hd+1）?
????????????????h=i;?
????????????end?
????????end?
????????for?i=1:m?
????????????for?j=1:d+1?
????????????????x（hij）=allbetter（1ij）;????????????%更新第h個組織的每個粒子位置為組織最優位置?
????????????????v（hij）=allbetter（2ij）;????????????%更新第h個組織的每個粒子速度為組織最優速度?
??????????????