实战caffe多标签分类——汽车品牌与车辆外观(C++接口)[详细实现+数据集]

前言

很多地方我们都需要用到多标签分类,比如一张图片,上面有只蓝猫,另一张图片上面有一只黄狗,那么我们要识别的时候,就可以采用多标签分类这一思想了。任务一是识别出这个到底是猫还是狗?(类型)任务二是识别出这是蓝还是黄?(颜色)
这里写图片描述
这里写图片描述
网上看了几篇教程,有讲的非常好的,也有出bug飞上了天的(吐槽啊喂!)这里还是主要讲讲这篇:http://chuansong.me/n/494753151240。我自己已经测试了,可行,给薛大牛一个赞!但是遗憾的是这篇文章的内容严重不足啊(连lmdb生成的命令行格式都没有,还是我自己看代码琢磨了一下…)我就给这篇文章补充补充,给一些例子。

任务

我这里给出一个具体的任务咯,要求在以下图片中,识别出汽车品牌和车辆外形。汽车品牌分为:Benz/BMW/Audi 车辆外形分为:Sedan/SUV。这是一个只有72张图片的小数据库,包括了测试和训练集:
这里写图片描述
其中标注是这样的,Audi=0,BMW=1,Benz=2. Sedan =0, SUV=1。所以如果这辆车是奥迪的SUV,标注就是: xx.jpg 0 1。在数据库中,标注已经做好了。数据集的下载方式在文章的最后。

定义我们的网络结构

我们这里采用的是上述文章中薛大牛的方法,两个data层,一个data只放图片,另一个data放label,label通过slice layer切开。然后我们开始定义网络!修改AlexNet!这是我的网络:

name: "ZnNet"
layer {
  name: "data"
  type: "Data"
  top: "data"  
  transform_param {
    mirror: true
    crop_size: 227
    mean_file: "models/bvlc_alexnet/ZnCarTrainMean.binaryproto"
  }
  include {
    phase: TRAIN
  }
  data_param {
    source: "models/bvlc_alexnet/ZnCarTrainImage" 
    batch_size: 10
    backend: LMDB
  }
}
layer {
  name: "labels"
  type: "Data"
  top: "labels"
  include {
    phase: TRAIN
  }
  data_param {
    source: "models/bvlc_alexnet/ZnCarTrainLabel"
    batch_size: 10
    backend: LMDB
  }
}
layer {
  name: "data"
  type: "Data"
  top: "data"
  transform_param {
     crop_size: 227    
     mean_file: "models/bvlc_alexnet/ZnCarTestMean.binaryproto"
  } 
  include {
    phase: TEST
  }
  data_param {
    source: "models/bvlc_alexnet/ZnCarTestImage"
    batch_size: 12
    backend: LMDB
   }
}
layer {
  name: "labels"
  type: "Data"
  top: "labels"
  include {
     phase: TEST
  }
  data_param {
    source: "models/bvlc_alexnet/ZnCarTestLabel"
    batch_size: 12
    backend: LMDB
  }
}
layer {
  name: "slice"
  type: "Slice"
  bottom: "labels"
  top: "type" #汽车品牌
  top: "surface" #车的外形
  slice_param {
    axis: 1
    slice_point: 1
  }
}
layer {
  name: "conv1"
  type: "Convolution"
  bottom: "data"
  top: "conv1"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 96
    kernel_size: 11
    stride: 4
    weight_filler {
      type: "gaussian"
      std: 0.01
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
}
layer {
  name: "relu1"
  type: "ReLU"
  bottom: "conv1"
  top: "conv1"
}
layer {
  name: "norm1"
  type: "LRN"
  bottom: "conv1"
  top: "norm1"
  lrn_param {
    local_size: 5
    alpha: 0.0001
    beta: 0.75
  }
}
layer {
  name: "pool1"
  type: "Pooling"
  bottom: "norm1"
  top: "pool1"
  pooling_param {
    pool: MAX
    kernel_size: 3
    stride: 2
  }
}
layer {
  name: "conv2"
  type: "Convolution"
  bottom: "pool1"
  top: "conv2"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 256
    pad: 2
    kernel_size: 5
    group: 2
    weight_filler {
      type: "gaussian"
      std: 0.01
    }
    bias_filler {
      type: "constant"
      value: 0.1
    }
  }
}
layer {
  name: "relu2"
  type: "ReLU"
  bottom: "conv2"
  top: "conv2"
}
layer {
  name: "norm2"
  type: "LRN"
  bottom: "conv2"
  top: "norm2"
  lrn_param {
    local_size: 5
    alpha: 0.0001
    beta: 0.75
  }
}
layer {
  name: "pool2"
  type: "Pooling"
  bottom: "norm2"
  top: "pool2"
  pooling_param {
    pool: MAX
    kernel_size: 3
    stride: 2
    }
}
layer {
  name: "conv3"
  type: "Convolution"
  bottom: "pool2"
  top: "conv3"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 384
    pad: 1
    kernel_size: 3
    weight_filler {
      type: "gaussian"
      std: 0.01
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
}
layer {
  name: "relu3"
  type: "ReLU"
  bottom: "conv3"
  top: "conv3"
}
layer {
  name: "conv4"
  type: "Convolution"
  bottom: "conv3"
  top: "conv4"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 384
    pad: 1
    kernel_size: 3
    group: 2
    weight_filler {
      type: "gaussian"
      std: 0.01
    }
    bias_filler {
      type: "constant"
      value: 0.1
    }
  }
}
layer {
  name: "relu4"
  type: "ReLU"
  bottom: "conv4"
  top: "conv4"
}
layer {
  name: "conv5"
  type: "Convolution"
  bottom: "conv4"
  top: "conv5"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 256
    pad: 1
    kernel_size: 3
    group: 2
    weight_filler {
      type: "gaussian"
      std: 0.01
    }
    bias_filler {
      type: "constant"
      value: 0.1
    }
  }
}
layer {
  name: "relu5"
  type: "ReLU"
  bottom: "conv5"
  top: "conv5"
}
layer {
  name: "pool5"
  type: "Pooling"
  bottom: "conv5"
  top: "pool5"
  pooling_param {
    pool: MAX
    kernel_size: 3
    stride: 2
  }
}
layer {
  name: "fc6"
  type: "InnerProduct"
  bottom: "pool5"
  top: "fc6"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  inner_product_param {
    num_output: 4096
    weight_filler {
      type: "gaussian"
      std: 0.005
    }
    bias_filler {
      type: "constant"
      value: 0.1
    }
  }
}
layer {
  name: "relu6"
  type: "ReLU"
  bottom: "fc6"
  top: "fc6"
}
layer {
  name: "drop6"
  type: "Dropout"
  bottom: "fc6"
  top: "fc6"
  dropout_param {
    dropout_ratio: 0.5
  }
}
layer {
  name: "fc7"
  type: "InnerProduct"
  bottom: "fc6"
  top: "fc7"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  inner_product_param {
    num_output: 4096
    weight_filler {
      type: "gaussian"
      std: 0.005
    }
    bias_filler {
      type: "constant"
      value: 0.1
    }
  }
}
layer {
  name: "relu7"
  type: "ReLU"
  bottom: "fc7"
  top: "fc7"
}
layer {
  name: "drop7"
  type: "Dropout"
  bottom: "fc7"
  top: "fc7"
  dropout_param {
    dropout_ratio: 0.5
  }
}
layer {
  name: "fc8_type"
  type: "InnerProduct"
  bottom: "fc7"
  top: "fc8_type"
  param {
    lr_mult: 5
    decay_mult: 5
  }
  param {
    lr_mult: 10
    decay_mult: 0
  }
  inner_product_param {
    num_output: 3
    weight_filler {
      type: "gaussian"
      std: 0.01
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
}
layer {
  name: "fc8_surface"
  type: "InnerProduct"
  bottom: "fc7"
  top: "fc8_surface"
  param {
    lr_mult: 5
    decay_mult: 5
  }
  param {
    lr_mult: 10
    decay_mult: 0
  }
  inner_product_param {
    num_output: 2
    weight_filler {
      type: "gaussian"
      std: 0.01
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
}
layer {
  name: "accuracy_type"
  type: "Accuracy"
  bottom: "fc8_type"
  bottom: "type"
  top: "accuracy_type"
  include {
    phase: TEST
  }
}
layer {
  name: "loss_type"
  type: "SoftmaxWithLoss"
  bottom: "fc8_type"
  bottom: "type"
  top: "loss_type"
  loss_weight:0.5
}
layer {
  name: "accuracy_surface"
  type: "Accuracy"
  bottom: "fc8_surface"
  bottom: "surface"
  top: "accuracy_surface"
  include {
    phase: TEST
  }
}
layer {
  name: "loss_surface"
  type: "SoftmaxWithLoss"
  bottom: "fc8_surface"
  bottom: "surface"
  top: "loss_surface"
  loss_weight:0.5
}
  • 1
  • 2
  • 3
  • 4
  • 5
  • 6
  • 7
  • 8
  • 9
  • 10
  • 11
  • 12
  • 13
  • 14
  • 15
  • 16
  • 17
  • 18
  • 19
  • 20
  • 21
  • 22
  • 23
  • 24
  • 25
  • 26
  • 27
  • 28
  • 29
  • 30
  • 31
  • 32
  • 33
  • 34
  • 35
  • 36
  • 37
  • 38
  • 39
  • 40
  • 41
  • 42
  • 43
  • 44
  • 45
  • 46
  • 47
  • 48
  • 49
  • 50
  • 51
  • 52
  • 53
  • 54
  • 55
  • 56
  • 57
  • 58
  • 59
  • 60
  • 61
  • 62
  • 63
  • 64
  • 65
  • 66
  • 67
  • 68
  • 69
  • 70
  • 71
  • 72
  • 73
  • 74
  • 75
  • 76
  • 77
  • 78
  • 79
  • 80
  • 81
  • 82
  • 83
  • 84
  • 85
  • 86
  • 87
  • 88
  • 89
  • 90
  • 91
  • 92
  • 93
  • 94
  • 95
  • 96
  • 97
  • 98
  • 99
  • 100
  • 101
  • 102
  • 103
  • 104
  • 105
  • 106
  • 107
  • 108
  • 109
  • 110
  • 111
  • 112
  • 113
  • 114
  • 115
  • 116
  • 117
  • 118
  • 119
  • 120
  • 121
  • 122
  • 123
  • 124
  • 125
  • 126
  • 127
  • 128
  • 129
  • 130
  • 131
  • 132
  • 133
  • 134
  • 135
  • 136
  • 137
  • 138
  • 139
  • 140
  • 141
  • 142
  • 143
  • 144
  • 145
  • 146
  • 147
  • 148
  • 149
  • 150
  • 151
  • 152
  • 153
  • 154
  • 155
  • 156
  • 157
  • 158
  • 159
  • 160
  • 161
  • 162
  • 163
  • 164
  • 165
  • 166
  • 167
  • 168
  • 169
  • 170
  • 171
  • 172
  • 173
  • 174
  • 175
  • 176
  • 177
  • 178
  • 179
  • 180
  • 181
  • 182
  • 183
  • 184
  • 185
  • 186
  • 187
  • 188
  • 189
  • 190
  • 191
  • 192
  • 193
  • 194
  • 195
  • 196
  • 197
  • 198
  • 199
  • 200
  • 201
  • 202
  • 203
  • 204
  • 205
  • 206
  • 207
  • 208
  • 209
  • 210
  • 211
  • 212
  • 213
  • 214
  • 215
  • 216
  • 217
  • 218
  • 219
  • 220
  • 221
  • 222
  • 223
  • 224
  • 225
  • 226
  • 227
  • 228
  • 229
  • 230
  • 231
  • 232
  • 233
  • 234
  • 235
  • 236
  • 237
  • 238
  • 239
  • 240
  • 241
  • 242
  • 243
  • 244
  • 245
  • 246
  • 247
  • 248
  • 249
  • 250
  • 251
  • 252
  • 253
  • 254
  • 255
  • 256
  • 257
  • 258
  • 259
  • 260
  • 261
  • 262
  • 263
  • 264
  • 265
  • 266
  • 267
  • 268
  • 269
  • 270
  • 271
  • 272
  • 273
  • 274
  • 275
  • 276
  • 277
  • 278
  • 279
  • 280
  • 281
  • 282
  • 283
  • 284
  • 285
  • 286
  • 287
  • 288
  • 289
  • 290
  • 291
  • 292
  • 293
  • 294
  • 295
  • 296
  • 297
  • 298
  • 299
  • 300
  • 301
  • 302
  • 303
  • 304
  • 305
  • 306
  • 307
  • 308
  • 309
  • 310
  • 311
  • 312
  • 313
  • 314
  • 315
  • 316
  • 317
  • 318
  • 319
  • 320
  • 321
  • 322
  • 323
  • 324
  • 325
  • 326
  • 327
  • 328
  • 329
  • 330
  • 331
  • 332
  • 333
  • 334
  • 335
  • 336
  • 337
  • 338
  • 339
  • 340
  • 341
  • 342
  • 343
  • 344
  • 345
  • 346
  • 347
  • 348
  • 349
  • 350
  • 351
  • 352
  • 353
  • 354
  • 355
  • 356
  • 357
  • 358
  • 359
  • 360
  • 361
  • 362
  • 363
  • 364
  • 365
  • 366
  • 367
  • 368
  • 369
  • 370
  • 371
  • 372
  • 373
  • 374
  • 375
  • 376
  • 377
  • 378
  • 379
  • 380
  • 381
  • 382
  • 383
  • 384
  • 385
  • 386
  • 387
  • 388
  • 389
  • 390
  • 391
  • 392
  • 393
  • 394
  • 395
  • 396
  • 397
  • 398
  • 399
  • 400
  • 401
  • 402
  • 403
  • 404
  • 405
  • 406
  • 407
  • 408
  • 409
  • 410
  • 411
  • 412
  • 413
  • 414
  • 415
  • 416
  • 417
  • 418
  • 419
  • 420
  • 421
  • 422
  • 423
  • 424
  • 425
  • 426
  • 427
  • 428
  • 429
  • 430
  • 431
  • 432
  • 433
  • 434
  • 435
  • 436
  • 437
  • 438
  • 439
  • 440
  • 441
  • 442
  • 443
  • 444
  • 445
  • 446
  • 447
  • 448
  • 449
  • 450
  • 451
  • 452
  • 453
  • 454
  • 455
  • 456
  • 457
  • 458
  • 459
  • 460
  • 461
  • 462

loss_weight是指的这一层的loss对整个网络反向传播时的贡献。我们这里两个loss,先各自设定0.5。在这里推荐大家一个网站:http://ethereon.github.io/netscope/#/editor输入自己的网络定义文件,输出直观的网络图。这样一来,我们就可以很直观的看啦:
这里写图片描述
label被slice分成了两个层,各自对应相关的loss。(看起来还是很合理的对吧?)

做数据转换工具!

整个网络结构需要两类数据,一类是纯图片的lmdb,一类是包含两个标签的lmdb。首先我们需要修改caffe中的源码(相信我,很简单!)打开你编译caffe时候的工程,找到convert_imageset这个工程:
这里写图片描述
修改convert_imageset.cpp为convert_multilabel.cpp,内容如下,全部替代也可以:(见github:https://github.com/HolidayXue/CodeSnap/blob/master/convert_multilabel.cpp)然后重新编译这个convert_imageset这个project。你可能会遇到什么没有生成object的警告,在这里关掉即可:
这里写图片描述
完成后,我们会看到release文件夹下多了一个convert_multilabel的exe,这个就是我们的lmdb生成工具啦。

做图像和多标签数据!

下载我的(或者自己做也可以)数据集,用命令行命令:

convert_multilabel.exe  --resize_height=227  --resize_width=227 ZnCarTrain/ ZnCarTrain/Label.txt ZnCarTrainImage ZnCarTrainLabel 2
  • 1

几个参数我解释一下。–resize_height=227 –resize_width=227 代表将图像缩放到227*227,ZnCar/是目录,ZnCarTrain/Label.txt是你的标注所在的地方,ZnCarTrainImage和ZnCarTrainLabel是要生成的lmdb文件夹,最后一个2代表着你这里有两类标签。(如果有更多标签需要分类的话,这里要改,slice layer和前面的loss都需要改哦~)
然后还要做均值文件:

compute_image_mean.exe convert_data_train image_mean.binaryproto
pause
  • 1
  • 2

总之最后应该有六个东西,如图:
这里写图片描述
数据制作完毕。

微调AlexNet!

网上下载一个bvlc_alexnet.caffemodel的权重文件。然后就是finetune了,参数的话得好好改改solver.prototxt里面,比如base_lr调小一点啦,迭代max_iter不要太多啦等等。此外也可以把两个fc8层的学习率适当提升一些,因为在finetune的时候训练好的权重只对原来没变的层有作用,这个新层就相当于是随机初始化的参数。
我这是用cpu训练的,很慢。(早知道就把显卡带回来了):
这里写图片描述

试一试我们的模型

我们做一个deploy文件:

name: "ZnNet"
layer {
  name: "data"
  type: "Input"
  top: "data"
  input_param { shape: { dim: 1 dim: 3 dim: 227 dim: 227 } }
}
layer {
  name: "conv1"
  type: "Convolution"
  bottom: "data"
  top: "conv1"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 96
    kernel_size: 11
    stride: 4
    weight_filler {
      type: "gaussian"
      std: 0.01
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
}
layer {
  name: "relu1"
  type: "ReLU"
  bottom: "conv1"
  top: "conv1"
}
layer {
  name: "norm1"
  type: "LRN"
  bottom: "conv1"
  top: "norm1"
  lrn_param {
    local_size: 5
    alpha: 0.0001
    beta: 0.75
  }
}
layer {
  name: "pool1"
  type: "Pooling"
  bottom: "norm1"
  top: "pool1"
  pooling_param {
    pool: MAX
    kernel_size: 3
    stride: 2
  }
}
layer {
  name: "conv2"
  type: "Convolution"
  bottom: "pool1"
  top: "conv2"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 256
    pad: 2
    kernel_size: 5
    group: 2
    weight_filler {
      type: "gaussian"
      std: 0.01
    }
    bias_filler {
      type: "constant"
      value: 0.1
    }
  }
}
layer {
  name: "relu2"
  type: "ReLU"
  bottom: "conv2"
  top: "conv2"
}
layer {
  name: "norm2"
  type: "LRN"
  bottom: "conv2"
  top: "norm2"
  lrn_param {
    local_size: 5
    alpha: 0.0001
    beta: 0.75
  }
}
layer {
  name: "pool2"
  type: "Pooling"
  bottom: "norm2"
  top: "pool2"
  pooling_param {
    pool: MAX
    kernel_size: 3
    stride: 2
    }
}
layer {
  name: "conv3"
  type: "Convolution"
  bottom: "pool2"
  top: "conv3"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 384
    pad: 1
    kernel_size: 3
    weight_filler {
      type: "gaussian"
      std: 0.01
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
}
layer {
  name: "relu3"
  type: "ReLU"
  bottom: "conv3"
  top: "conv3"
}
layer {
  name: "conv4"
  type: "Convolution"
  bottom: "conv3"
  top: "conv4"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 384
    pad: 1
    kernel_size: 3
    group: 2
    weight_filler {
      type: "gaussian"
      std: 0.01
    }
    bias_filler {
      type: "constant"
      value: 0.1
    }
  }
}
layer {
  name: "relu4"
  type: "ReLU"
  bottom: "conv4"
  top: "conv4"
}
layer {
  name: "conv5"
  type: "Convolution"
  bottom: "conv4"
  top: "conv5"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 256
    pad: 1
    kernel_size: 3
    group: 2
    weight_filler {
      type: "gaussian"
      std: 0.01
    }
    bias_filler {
      type: "constant"
      value: 0.1
    }
  }
}
layer {
  name: "relu5"
  type: "ReLU"
  bottom: "conv5"
  top: "conv5"
}
layer {
  name: "pool5"
  type: "Pooling"
  bottom: "conv5"
  top: "pool5"
  pooling_param {
    pool: MAX
    kernel_size: 3
    stride: 2
  }
}
layer {
  name: "fc6"
  type: "InnerProduct"
  bottom: "pool5"
  top: "fc6"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  inner_product_param {
    num_output: 4096
    weight_filler {
      type: "gaussian"
      std: 0.005
    }
    bias_filler {
      type: "constant"
      value: 0.1
    }
  }
}
layer {
  name: "relu6"
  type: "ReLU"
  bottom: "fc6"
  top: "fc6"
}
layer {
  name: "drop6"
  type: "Dropout"
  bottom: "fc6"
  top: "fc6"
  dropout_param {
    dropout_ratio: 0.5
  }
}
layer {
  name: "fc7"
  type: "InnerProduct"
  bottom: "fc6"
  top: "fc7"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  inner_product_param {
    num_output: 4096
    weight_filler {
      type: "gaussian"
      std: 0.005
    }
    bias_filler {
      type: "constant"
      value: 0.1
    }
  }
}
layer {
  name: "relu7"
  type: "ReLU"
  bottom: "fc7"
  top: "fc7"
}
layer {
  name: "drop7"
  type: "Dropout"
  bottom: "fc7"
  top: "fc7"
  dropout_param {
    dropout_ratio: 0.5
  }
}
layer {
  name: "fc8_type"
  type: "InnerProduct"
  bottom: "fc7"
  top: "fc8_type"
  param {
    lr_mult: 5
    decay_mult: 5
  }
  param {
    lr_mult: 10
    decay_mult: 0
  }
  inner_product_param {
    num_output: 3
    weight_filler {
      type: "gaussian"
      std: 0.01
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
}
layer {
  name: "fc8_surface"
  type: "InnerProduct"
  bottom: "fc7"
  top: "fc8_surface"
  param {
    lr_mult: 5
    decay_mult: 5
  }
  param {
    lr_mult: 10
    decay_mult: 0
  }
  inner_product_param {
    num_output: 2
    weight_filler {
      type: "gaussian"
      std: 0.01
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
}
layer {
  name: "prob_type"
  type: "Softmax"
  bottom: "fc8_type"
  top: "prob_type"
  loss_weight:0.5
}
layer {
  name: "prob_surface"
  type: "Softmax"
  bottom: "fc8_surface"
  top: "prob_surface"
  loss_weight:0.5
}
  • 1
  • 2
  • 3
  • 4
  • 5
  • 6
  • 7
  • 8
  • 9
  • 10
  • 11
  • 12
  • 13
  • 14
  • 15
  • 16
  • 17
  • 18
  • 19
  • 20
  • 21
  • 22
  • 23
  • 24
  • 25
  • 26
  • 27
  • 28
  • 29
  • 30
  • 31
  • 32
  • 33
  • 34
  • 35
  • 36
  • 37
  • 38
  • 39
  • 40
  • 41
  • 42
  • 43
  • 44
  • 45
  • 46
  • 47
  • 48
  • 49
  • 50
  • 51
  • 52
  • 53
  • 54
  • 55
  • 56
  • 57
  • 58
  • 59
  • 60
  • 61
  • 62
  • 63
  • 64
  • 65
  • 66
  • 67
  • 68
  • 69
  • 70
  • 71
  • 72
  • 73
  • 74
  • 75
  • 76
  • 77
  • 78
  • 79
  • 80
  • 81
  • 82
  • 83
  • 84
  • 85
  • 86
  • 87
  • 88
  • 89
  • 90
  • 91
  • 92
  • 93
  • 94
  • 95
  • 96
  • 97
  • 98
  • 99
  • 100
  • 101
  • 102
  • 103
  • 104
  • 105
  • 106
  • 107
  • 108
  • 109
  • 110
  • 111
  • 112
  • 113
  • 114
  • 115
  • 116
  • 117
  • 118
  • 119
  • 120
  • 121
  • 122
  • 123
  • 124
  • 125
  • 126
  • 127
  • 128
  • 129
  • 130
  • 131
  • 132
  • 133
  • 134
  • 135
  • 136
  • 137
  • 138
  • 139
  • 140
  • 141
  • 142
  • 143
  • 144
  • 145
  • 146
  • 147
  • 148
  • 149
  • 150
  • 151
  • 152
  • 153
  • 154
  • 155
  • 156
  • 157
  • 158
  • 159
  • 160
  • 161
  • 162
  • 163
  • 164
  • 165
  • 166
  • 167
  • 168
  • 169
  • 170
  • 171
  • 172
  • 173
  • 174
  • 175
  • 176
  • 177
  • 178
  • 179
  • 180
  • 181
  • 182
  • 183
  • 184
  • 185
  • 186
  • 187
  • 188
  • 189
  • 190
  • 191
  • 192
  • 193
  • 194
  • 195
  • 196
  • 197
  • 198
  • 199
  • 200
  • 201
  • 202
  • 203
  • 204
  • 205
  • 206
  • 207
  • 208
  • 209
  • 210
  • 211
  • 212
  • 213
  • 214
  • 215
  • 216
  • 217
  • 218
  • 219
  • 220
  • 221
  • 222
  • 223
  • 224
  • 225
  • 226
  • 227
  • 228
  • 229
  • 230
  • 231
  • 232
  • 233
  • 234
  • 235
  • 236
  • 237
  • 238
  • 239
  • 240
  • 241
  • 242
  • 243
  • 244
  • 245
  • 246
  • 247
  • 248
  • 249
  • 250
  • 251
  • 252
  • 253
  • 254
  • 255
  • 256
  • 257
  • 258
  • 259
  • 260
  • 261
  • 262
  • 263
  • 264
  • 265
  • 266
  • 267
  • 268
  • 269
  • 270
  • 271
  • 272
  • 273
  • 274
  • 275
  • 276
  • 277
  • 278
  • 279
  • 280
  • 281
  • 282
  • 283
  • 284
  • 285
  • 286
  • 287
  • 288
  • 289
  • 290
  • 291
  • 292
  • 293
  • 294
  • 295
  • 296
  • 297
  • 298
  • 299
  • 300
  • 301
  • 302
  • 303
  • 304
  • 305
  • 306
  • 307
  • 308
  • 309
  • 310
  • 311
  • 312
  • 313
  • 314
  • 315
  • 316
  • 317
  • 318
  • 319
  • 320
  • 321
  • 322
  • 323
  • 324
  • 325
  • 326
  • 327
  • 328
  • 329
  • 330
  • 331
  • 332
  • 333
  • 334
  • 335
  • 336
  • 337
  • 338
  • 339
  • 340
  • 341
  • 342
  • 343
  • 344
  • 345
  • 346
  • 347
  • 348
  • 349
  • 350
  • 351
  • 352
  • 353
  • 354
  • 355
  • 356
  • 357
  • 358
  • 359
  • 360
  • 361
  • 362
  • 363
  • 364
  • 365
  • 366
  • 367
  • 368
  • 369
  • 370
  • 371
  • 372
  • 373
  • 374
  • 375

然后这里还需要修改classification.cpp文件,因为Caffe自带的只支持单样本。我自己做了一个修改版本的,主要是修改了命令行和输出网络的部分,这个cpp只支持两标签,主要是用着方便,如果你需要更多的标签可以自己修改。编译步骤同convert_multilabel.cpp的编译。底下的代码直接就可以替换原来的cpp。
convert_multilabel.cpp:

#include <caffe/caffe.hpp>
#ifdef USE_OPENCV
#include <opencv2/core/core.hpp>
#include <opencv2/highgui/highgui.hpp>
#include <opencv2/imgproc/imgproc.hpp>
#endif  // USE_OPENCV
#include <algorithm>
#include <iosfwd>
#include <memory>
#include <string>
#include <utility>
#include <vector>

#ifdef USE_OPENCV
using namespace caffe;  // NOLINT(build/namespaces)
using std::string;

/* Pair (label, confidence) representing a prediction. */
// change: 2 classify — (label1, confidence1) (label2, confidence2)
typedef std::pair<string, float> Prediction;
class Classifier {
 public:
  Classifier(const string& model_file,
             const string& trained_file,
             const string& mean_file,
             const vector<string>& label_files);

  std::vector<vector<Prediction>> Classify(const cv::Mat& img, int N = 5);

 private:
  void SetMean(const string& mean_file);

  std::vector<vector<float>> Predict(const cv::Mat& img);

  void WrapInputLayer(std::vector<cv::Mat>* input_channels);

  void Preprocess(const cv::Mat& img,
                  std::vector<cv::Mat>* input_channels);

 private:
  shared_ptr<Net<float> > net_;
  cv::Size input_geometry_;
  int num_channels_;
  cv::Mat mean_;
  std::vector<vector<string>> labels_; //multi
};

Classifier::Classifier(const string& model_file,
                       const string& trained_file,
                       const string& mean_file,
                       const vector<string>& label_file) {
#ifdef CPU_ONLY
  Caffe::set_mode(Caffe::CPU);
#else
  Caffe::set_mode(Caffe::GPU);
#endif

  /* Load the network. */
  net_.reset(new Net<float>(model_file, TEST));
  net_->CopyTrainedLayersFrom(trained_file);

  CHECK_EQ(net_->num_inputs(), 1) << "Network should have exactly one input.";
  //CHECK_EQ(net_->num_outputs(), 1) << "Network should have exactly one output.";

  Blob<float>* input_layer = net_->input_blobs()[0];
  num_channels_ = input_layer->channels();
  CHECK(num_channels_ == 3 || num_channels_ == 1)
    << "Input layer should have 1 or 3 channels.";
  input_geometry_ = cv::Size(input_layer->width(), input_layer->height());

  /* Load the binaryproto mean file. */
  SetMean(mean_file);

  /* Load labels. */
  //2 labels should read
  string line;
  for (int i = 0; i < label_file.size(); i++)
  {
      std::ifstream labels(label_file[i].c_str());
      CHECK(labels) << "Unable to open labels file " << label_file[i];
      vector<string> label_array;
      while (std::getline(labels, line))
      {
          label_array.push_back(line);
      }
      Blob<float>* output_layer = net_->output_blobs()[i];
      CHECK_EQ(label_array.size(), output_layer->channels())
          << "Number of labels is different from the output layer dimension.";
      labels_.push_back(label_array);
  }
}

static bool PairCompare(const std::pair<float, int>& lhs,
                        const std::pair<float, int>& rhs) {
  return lhs.first > rhs.first;
}

/* Return the indices of the top N values of vector v. */
static std::vector<int> Argmax(const std::vector<float>& v, int N) {
  std::vector<std::pair<float, int> > pairs;
  for (size_t i = 0; i < v.size(); ++i)
    pairs.push_back(std::make_pair(v[i], static_cast<int>(i)));
  std::partial_sort(pairs.begin(), pairs.begin() + N, pairs.end(), PairCompare);

  std::vector<int> result;
  for (int i = 0; i < N; ++i)
    result.push_back(pairs[i].second);
  return result;
}

/* Return the top N predictions. */
std::vector<vector<Prediction>> Classifier::Classify(const cv::Mat& img, int N) {
  auto output = Predict(img);
  int N1 = std::min<int>(labels_[0].size(), N);
  int N2 = std::min<int>(labels_[1].size(), N);
  std::vector<int> maxN1 = Argmax(output[0], N1);
  std::vector<int> maxN2 = Argmax(output[1], N2);
  std::vector<Prediction> predictions1;
  std::vector<Prediction> predictions2;

  for (int i = 0; i < N1; ++i) {
    int idx = maxN1[i];
    predictions1.push_back(std::make_pair(labels_[0][idx], output[0][idx]));
  }
  for (int i = 0; i < N2; ++i) {
      int idx = maxN2[i];
      predictions2.push_back(std::make_pair(labels_[1][idx], output[1][idx]));
  }
  vector<vector<Prediction>> predictions;
  predictions.push_back(predictions1);
  predictions.push_back(predictions2);
  return predictions;
}

/* Load the mean file in binaryproto format. */
void Classifier::SetMean(const string& mean_file) {
  BlobProto blob_proto;
  ReadProtoFromBinaryFileOrDie(mean_file.c_str(), &blob_proto);

  /* Convert from BlobProto to Blob<float> */
  Blob<float> mean_blob;
  mean_blob.FromProto(blob_proto);
  CHECK_EQ(mean_blob.channels(), num_channels_)
    << "Number of channels of mean file doesn't match input layer.";

  /* The format of the mean file is planar 32-bit float BGR or grayscale. */
  std::vector<cv::Mat> channels;
  float* data = mean_blob.mutable_cpu_data();
  for (int i = 0; i < num_channels_; ++i) {
    /* Extract an individual channel. */
    cv::Mat channel(mean_blob.height(), mean_blob.width(), CV_32FC1, data);
    channels.push_back(channel);
    data += mean_blob.height() * mean_blob.width();
  }

  /* Merge the separate channels into a single image. */
  cv::Mat mean;
  cv::merge(channels, mean);

  /* Compute the global mean pixel value and create a mean image
   * filled with this value. */
  cv::Scalar channel_mean = cv::mean(mean);
  mean_ = cv::Mat(input_geometry_, mean.type(), channel_mean);
}

std::vector<vector<float>> Classifier::Predict(const cv::Mat& img) {
  Blob<float>* input_layer = net_->input_blobs()[0];
  input_layer->Reshape(1, num_channels_,
                       input_geometry_.height, input_geometry_.width);
  /* Forward dimension change to all layers. */
  net_->Reshape();

  std::vector<cv::Mat> input_channels;
  WrapInputLayer(&input_channels);

  Preprocess(img, &input_channels);

  net_->Forward();

  /* Copy the output layer to a std::vector */
  Blob<float>* output_layer1 = net_->output_blobs()[0];
  Blob<float>* output_layer2 = net_->output_blobs()[1];
  const float* begin1 = output_layer1->cpu_data();
  const float* end1 = begin1+ output_layer1->channels();
  const float* begin2 = output_layer2->cpu_data();
  const float* end2 = begin2 + output_layer2->channels();

  std::vector<float> prob1(begin1, end1);
  std::vector<float> prob2(begin2, end2);
  vector<vector<float>> prob_matrix;
  prob_matrix.push_back(prob1);
  prob_matrix.push_back(prob2);
  return prob_matrix;
}

/* Wrap the input layer of the network in separate cv::Mat objects
 * (one per channel). This way we save one memcpy operation and we
 * don't need to rely on cudaMemcpy2D. The last preprocessing
 * operation will write the separate channels directly to the input
 * layer. */
void Classifier::WrapInputLayer(std::vector<cv::Mat>* input_channels) {
  Blob<float>* input_layer = net_->input_blobs()[0];

  int width = input_layer->width();
  int height = input_layer->height();
  float* input_data = input_layer->mutable_cpu_data();
  for (int i = 0; i < input_layer->channels(); ++i) {
    cv::Mat channel(height, width, CV_32FC1, input_data);
    input_channels->push_back(channel);
    input_data += width * height;
  }
}

void Classifier::Preprocess(const cv::Mat& img,
                            std::vector<cv::Mat>* input_channels) {
  /* Convert the input image to the input image format of the network. */
  cv::Mat sample;
  if (img.channels() == 3 && num_channels_ == 1)
    cv::cvtColor(img, sample, cv::COLOR_BGR2GRAY);
  else if (img.channels() == 4 && num_channels_ == 1)
    cv::cvtColor(img, sample, cv::COLOR_BGRA2GRAY);
  else if (img.channels() == 4 && num_channels_ == 3)
    cv::cvtColor(img, sample, cv::COLOR_BGRA2BGR);
  else if (img.channels() == 1 && num_channels_ == 3)
    cv::cvtColor(img, sample, cv::COLOR_GRAY2BGR);
  else
    sample = img;

  cv::Mat sample_resized;
  if (sample.size() != input_geometry_)
    cv::resize(sample, sample_resized, input_geometry_);
  else
    sample_resized = sample;

  cv::Mat sample_float;
  if (num_channels_ == 3)
    sample_resized.convertTo(sample_float, CV_32FC3);
  else
    sample_resized.convertTo(sample_float, CV_32FC1);

  cv::Mat sample_normalized;
  cv::subtract(sample_float, mean_, sample_normalized);

  /* This operation will write the separate BGR planes directly to the
   * input layer of the network because it is wrapped by the cv::Mat
   * objects in input_channels. */
  cv::split(sample_normalized, *input_channels);

  CHECK(reinterpret_cast<float*>(input_channels->at(0).data)
        == net_->input_blobs()[0]->cpu_data())
    << "Input channels are not wrapping the input layer of the network.";
}

int main(int argc, char** argv) {
  if (argc != 7) {
    std::cerr << "Usage: " << argv[0]
              << " deploy.prototxt network.caffemodel"
              << " mean.binaryproto label1.txt label2.txt img.jpg" << std::endl;
    return 1;
  }

  ::google::InitGoogleLogging(argv[0]);

  string model_file   = argv[1];
  string trained_file = argv[2];
  string mean_file    = argv[3];
  string label_file1   = argv[4];
  string label_file2   = argv[5];
  vector<string> label_file;
  label_file.push_back(label_file1);
  label_file.push_back(label_file2);
  std::cout << "the labels' channel:"<<label_file.size() << std::endl;
  Classifier classifier(model_file, trained_file, mean_file, label_file);

  string file = argv[6];

  std::cout << "---------- Prediction for "
            << file << " ----------" << std::endl;

  cv::Mat img = cv::imread(file, -1);
  CHECK(!img.empty()) << "Unable to decode image " << file;
  auto predictions = classifier.Classify(img);
  std::cout << "have runed classifier.Classify" << std::endl;
  /* Print the top N predictions. */

  std::cout << "---------- Surface------------" << std::endl;
  for (size_t i = 0; i < predictions[0].size(); ++i) {
    Prediction p = predictions[0][i];
    std::cout << std::fixed << std::setprecision(4) << p.second << " - \""
              << p.first << "\"" << std::endl;
  }
  std::cout << " ---------- Type------------" << std::endl;
  for (size_t i = 0; i < predictions[1].size(); ++i) {
      Prediction p = predictions[1][i];
      std::cout << std::fixed << std::setprecision(4) << p.second << " - \""
          << p.first << "\"" << std::endl;
  }
}
#else
int main(int argc, char** argv) {
  LOG(FATAL) << "This example requires OpenCV; compile with USE_OPENCV.";
}
#endif  // USE_OPENCV
  • 1
  • 2
  • 3
  • 4
  • 5
  • 6
  • 7
  • 8
  • 9
  • 10
  • 11
  • 12
  • 13
  • 14
  • 15
  • 16
  • 17
  • 18
  • 19
  • 20
  • 21
  • 22
  • 23
  • 24
  • 25
  • 26
  • 27
  • 28
  • 29
  • 30
  • 31
  • 32
  • 33
  • 34
  • 35
  • 36
  • 37
  • 38
  • 39
  • 40
  • 41
  • 42
  • 43
  • 44
  • 45
  • 46
  • 47
  • 48
  • 49
  • 50
  • 51
  • 52
  • 53
  • 54
  • 55
  • 56
  • 57
  • 58
  • 59
  • 60
  • 61
  • 62
  • 63
  • 64
  • 65
  • 66
  • 67
  • 68
  • 69
  • 70
  • 71
  • 72
  • 73
  • 74
  • 75
  • 76
  • 77
  • 78
  • 79
  • 80
  • 81
  • 82
  • 83
  • 84
  • 85
  • 86
  • 87
  • 88
  • 89
  • 90
  • 91
  • 92
  • 93
  • 94
  • 95
  • 96
  • 97
  • 98
  • 99
  • 100
  • 101
  • 102
  • 103
  • 104
  • 105
  • 106
  • 107
  • 108
  • 109
  • 110
  • 111
  • 112
  • 113
  • 114
  • 115
  • 116
  • 117
  • 118
  • 119
  • 120
  • 121
  • 122
  • 123
  • 124
  • 125
  • 126
  • 127
  • 128
  • 129
  • 130
  • 131
  • 132
  • 133
  • 134
  • 135
  • 136
  • 137
  • 138
  • 139
  • 140
  • 141
  • 142
  • 143
  • 144
  • 145
  • 146
  • 147
  • 148
  • 149
  • 150
  • 151
  • 152
  • 153
  • 154
  • 155
  • 156
  • 157
  • 158
  • 159
  • 160
  • 161
  • 162
  • 163
  • 164
  • 165
  • 166
  • 167
  • 168
  • 169
  • 170
  • 171
  • 172
  • 173
  • 174
  • 175
  • 176
  • 177
  • 178
  • 179
  • 180
  • 181
  • 182
  • 183
  • 184
  • 185
  • 186
  • 187
  • 188
  • 189
  • 190
  • 191
  • 192
  • 193
  • 194
  • 195
  • 196
  • 197
  • 198
  • 199
  • 200
  • 201
  • 202
  • 203
  • 204
  • 205
  • 206
  • 207
  • 208
  • 209
  • 210
  • 211
  • 212
  • 213
  • 214
  • 215
  • 216
  • 217
  • 218
  • 219
  • 220
  • 221
  • 222
  • 223
  • 224
  • 225
  • 226
  • 227
  • 228
  • 229
  • 230
  • 231
  • 232
  • 233
  • 234
  • 235
  • 236
  • 237
  • 238
  • 239
  • 240
  • 241
  • 242
  • 243
  • 244
  • 245
  • 246
  • 247
  • 248
  • 249
  • 250
  • 251
  • 252
  • 253
  • 254
  • 255
  • 256
  • 257
  • 258
  • 259
  • 260
  • 261
  • 262
  • 263
  • 264
  • 265
  • 266
  • 267
  • 268
  • 269
  • 270
  • 271
  • 272
  • 273
  • 274
  • 275
  • 276
  • 277
  • 278
  • 279
  • 280
  • 281
  • 282
  • 283
  • 284
  • 285
  • 286
  • 287
  • 288
  • 289
  • 290
  • 291
  • 292
  • 293
  • 294
  • 295
  • 296
  • 297
  • 298
  • 299
  • 300
  • 301
  • 302
  • 303
  • 304

结果

写两个标注文件:

label1.txt:
Sedan
SUV

label2.txt:
Audi
BMW
Benz
  • 1
  • 2
  • 3
  • 4
  • 5
  • 6
  • 7
  • 8

我修改后的classification.exe 分类的命令行:

classification.exe deploy.prototxt network.caffemodel mean.binaryproto label1.txt label2.txt img.jpg
  • 1

这里写图片描述
这里写图片描述

本项目的Github网址:https://github.com/ChenJoya/Caffe_MultiLabel_Classification
数据集:http://download.csdn.net/detail/mr_curry/9742578
求星星~~

阅读更多 登录后自动展开
想对作者说点什么? 我来说一句

没有更多推荐了,返回首页