caffe 红绿灯识别

#coding=utf-8  
#加载必要的库  
import numpy as np  
 
import sys,os  
 
#设置当前目录  
caffe_root = '/home/ubuntu/caffe/'   
sys.path.insert(0, caffe_root + 'python')  
import caffe  
os.chdir(caffe_root)  
 
net_file='/home/ubuntu/Downloads/deep-learning-traffic-lights-master/model/deploy.prototxt'  
caffe_model='/home/ubuntu/Downloads/deep-learning-traffic-lights-master/model/train_squeezenet_scratch_trainval_manual_p2__iter_8000.caffemodel'  
mean_file=caffe_root + 'python/caffe/imagenet/ilsvrc_2012_mean.npy'  
 
net = caffe.Net(net_file,caffe_model,caffe.TEST)  
transformer = caffe.io.Transformer({'data': net.blobs['data'].data.shape})  
transformer.set_transpose('data', (2,0,1))  
transformer.set_mean('data', np.load(mean_file).mean(1).mean(1))  
transformer.set_raw_scale('data', 255)   
transformer.set_channel_swap('data', (2,1,0))  
 
im=caffe.io.load_image('/home/ubuntu/Downloads/deep-learning-traffic-lights-master/4.jpg')  
net.blobs['data'].data[...] = transformer.preprocess('data',im)  
out = net.forward()  
 
 
imagenet_labels_filename = '/home/ubuntu/Downloads/deep-learning-traffic-lights-master/synset_words.txt'
labels = np.loadtxt(imagenet_labels_filename, str, delimiter='\t')  
 
top_k = net.blobs['prob'].data[0].flatten().argsort()[-1:-6:-1]  
for i in np.arange(top_k.size):  
    print top_k[i], labels[top_k[i]]
#coding=utf-8  
#加载必要的库  
import numpy as np