caffe 红绿灯识别

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

[python] view plain copy

[python] view plain copy

synset_words.txt

1. 
   
2. yello
3. red
4. green