MATLAB神经网络图像识别高识别率代码

I0=pretreatment(imread('Z:\data\PictureData\TestCode\SplitDataTest\0 (1).png'));
I1=pretreatment(imread('Z:\data\PictureData\TestCode\SplitDataTest\1 (1).png'));
I2=pretreatment(imread('Z:\data\PictureData\TestCode\SplitDataTest\2 (1).png'));
I3=pretreatment(imread('Z:\data\PictureData\TestCode\SplitDataTest\3 (1).png'));
I4=pretreatment(imread('Z:\data\PictureData\TestCode\SplitDataTest\4 (1).png'));
I5=pretreatment(imread('Z:\data\PictureData\TestCode\SplitDataTest\5 (1).png'));
I6=pretreatment(imread('Z:\data\PictureData\TestCode\SplitDataTest\6 (1).png'));
I7=pretreatment(imread('Z:\data\PictureData\TestCode\SplitDataTest\7 (1).png'));
I8=pretreatment(imread('Z:\data\PictureData\TestCode\SplitDataTest\8 (1).png'));
I9=pretreatment(imread('Z:\data\PictureData\TestCode\SplitDataTest\9 (1).png'));
%以上数据都是归一化好的数据。
P=[I0',I1',I2',I3',I4',I5',I6',I7',I8',I9'];
T=eye(10,10);
%%bp神经网络参数设置
net=newff(minmax(P),[144,200,10],{'logsig','logsig','logsig'},'trainrp');
net.inputWeights{1,1}.initFcn ='randnr';
net.layerWeights{2,1}.initFcn ='randnr';
net.trainparam.epochs=5000;
net.trainparam.show=50;
net.trainparam.lr=0.001;
net.trainparam.goal=0.0000000000001;
net=init(net);
%%%训练样本%%%%
[net,tr]=train(net,P,T);
PIN0=pretreatment(imread('Z:\data\PictureData\TestCode\SplitDataTest\4 (2).png'));
PIN1=pretreatment(imread('Z:\data\PictureData\TestCode\SplitDataTest\3 (2).png'));
P0=[PIN0',PIN1'];
T0= sim(net ,PIN1')
T1 = compet (T0)
d =find(T1 == 1) - 1
fprintf('预测数字是:%d\n',d);
%有较高的识别率另外有需要云服务器可以了解下创新互联建站www.cdcxhl.com,海内外云服务器15元起步,三天无理由+7*72小时售后在线,公司持有idc许可证,提供“云服务器、裸金属服务器、高防服务器、香港服务器、美国服务器、虚拟主机、免备案服务器”等云主机租用服务以及企业上云的综合解决方案,具有“安全稳定、简单易用、服务可用性高、性价比高”等特点与优势,专为企业上云打造定制,能够满足用户丰富、多元化的应用场景需求。