跑通caffe-ssd demo代码(训练、测试自己数据集)

跑通caffe-ssd demo代码(训练、测试自己数据集)

ssd网络我就不多介绍了,CSDN上面一搜一大把。这篇主要讲讲如何跑通caffe代码~
github:caffe-ssd

一、代码结构

跑通caffe-ssd demo代码(训练、测试自己数据集)_第1张图片
在caffe-ssd中能用到的文件我全部在上述图片中标出来了,到时候具体的再细说~
caffe-ssd的环境自己百度吧,网上很多安装教程~

二、数据集准备

训练模型,首先第一个事情就是准备数据集。在利用caffe训练分类模型的时候,通常使用lmdb或者hdf5格式的数据,但是在该项目中使用的是lmdb格式的(其他格式的数据肯定也行,但是就是需要自己c++手写数据处理层了~太麻烦了)

1.

首先,准备VOC格式的数据
跑通caffe-ssd demo代码(训练、测试自己数据集)_第2张图片

这是VOC数据集的格式,其中,Annotations里面存放的是所有数据图片相对应的xml标签文件,imagesets里面存放的是一些txt文件,后续用到再详细说,JPEGImages里面存放的就是训练图片,最后两个文件夹是用于实例分割的,在目标检测中用不到,所以就不用管它。(这里,为了简单,所以只使用VOC_trainval_2007中前36张图片)
JPEGImages:
跑通caffe-ssd demo代码(训练、测试自己数据集)_第3张图片
其中,trainval里面存放训练图片,test存放测试图片。这个划分根据你自己项目进行决定就行。一般来说,trainval:test=4:1,这里我在test存放8张随机图片,剩余的存放在trainval文件夹中。值得一提,全部的图片还是放在trainval和test文件夹同级目录中,这样便于后续生成trainval.txt和test.txt。
Annotations:
跑通caffe-ssd demo代码(训练、测试自己数据集)_第4张图片
Imagesets:
跑通caffe-ssd demo代码(训练、测试自己数据集)_第5张图片

2.

下面介绍如何生成Imagesets里面的这四个txt文件~
先说trainval.txt和test.txt这两个文件。

#! /usr/bin/python  
# -*- coding:UTF-8 -*-  
 
import os, sys  
import glob  

trainval_dir = r"D:\voc\VOC_test\JPEGImages\trainval"  #训练集图片存放地址
test_dir = r"D:\voc\VOC_test\JPEGImages\test"  #测试图片存放地址
 
trainval_img_lists = glob.glob(trainval_dir + '/*.jpg') #如果你的图片是png格式的,只需要修改最后的后缀
trainval_img_names = []        
for item in trainval_img_lists:  
    temp1, temp2 = os.path.splitext(os.path.basename(item))  
    trainval_img_names.append(temp1)  
 
test_img_lists = glob.glob(test_dir + '/*.jpg') #如果你的图片是png格式的,只需要修改最后的后缀
test_img_names = []  
for item in test_img_lists:  
    temp1, temp2 = os.path.splitext(os.path.basename(item))  
    test_img_names.append(temp1)  
#图片路径和xml路径  
dist_img_dir = r"D:\voc\VOC_test\JPEGImages" #JPEGImages路径
dist_anno_dir = r"D:\voc\VOC_test\Annotations" #存放所有数据的xml文件路径
 
trainval_fd = open(r"D:\voc\VOC_test\ImageSets\trainval.txt", 'w')  #trainval.txt存放地址
test_fd = open(r"D:\voc\VOC_test\ImageSets\test.txt", 'w')  #test.txt存放地址
   
for item in trainval_img_names:  
    trainval_fd.write(dist_img_dir + '/' + str(item) + '.jpg' + ' ' + dist_anno_dir + '/' + str(item) + '.xml\n')  
 
for item in test_img_names:  
    test_fd.write(dist_img_dir + '/' + str(item) + '.jpg' + ' ' + dist_anno_dir + '/' + str(item) + '.xml\n')

生成的trainval.txt:
跑通caffe-ssd demo代码(训练、测试自己数据集)_第6张图片
test.txt:
跑通caffe-ssd demo代码(训练、测试自己数据集)_第7张图片
注意:这里我建议大家使用绝对路径,到时候在训练的时候比较清楚点

3.

下面介绍labelmap_voc.prototxt:
labelmap_voc.prototxt文件在你下载的caffe-ssd中有一个副本,位置在:you_caffe_root/data/VOC0712/里面,这里面的数据要根据你自己的需求进行修改,因为我这里就是VOC数据集,所以我不用改变的。不过假如我想进行猫狗目标检测算法,那我就得这么改:
跑通caffe-ssd demo代码(训练、测试自己数据集)_第8张图片
这里,label为0的是背景一类,不管你是检测多少种物体,这一类是不能动的。

4.

下面就是test_name_size.txt生成方式:

#! /usr/bin/python
# -*- coding:UTF-8 -*-
import os, sys
import glob
from PIL import Image #读图

img_dir = r"D:/voc/VOC_test/JPEGImages/test" #测试图片存放路径

#获取制定路径下的所有jpg图片的名称
img_lists = glob.glob(img_dir + '/*.jpg')

test_name_size = open(r"D:/voc/VOC_test/ImageSets/test_name_size.txt", 'w') #test_name_size.txt存放路径

for item in img_lists:
    img = Image.open(item)
    width, height = img.size
    temp1, temp2 = os.path.splitext(os.path.basename(item))
    test_name_size.write(temp1 + ' ' + str(height) + ' ' + str(width) + '\n')

最后生成的test_name_size.txt:
跑通caffe-ssd demo代码(训练、测试自己数据集)_第9张图片
其中,每一列中第一个表示测试图片名称,第二个和第三个表示的是该测试图片的高和宽。


这样,所有准备工作都做完了~用上述所有文件就可以生成lmdb数据了

5.

将VOC文件夹放在you_caffe_root/data中
跑通caffe-ssd demo代码(训练、测试自己数据集)_第10张图片
定位到you_caffe_root/data/VOC0712,下面应该有两个shell脚本文件:create_list.sh和create_data.sh。前者其实就是生成trainval.txt和test.txt,因为我们已经用python脚本生成好了,所以就可以直接用后者来生成lmdb数据了。
create_data.sh:

cur_dir=$(cd $( dirname ${BASH_SOURCE[0]} ) && pwd )
root_dir=$cur_dir/../.. #你安装caffe的根目录

cd $root_dir

redo=1
data_root_dir="" #上述VOC总文件夹的位置
mapfile="" #上述生成的labelmap_voc.prototxt存放位置
anno_type="detection"
db="lmdb"
min_dim=0
max_dim=0
width=0
height=0

extra_cmd="--encode-type=jpg --encoded"
if [ $redo ]
then
  extra_cmd="$extra_cmd --redo"
fi
for subset in test trainval
do
  python $root_dir/scripts/create_annoset.py #该sh脚本本质上是调用的是scripts/create_annoset.py,你找到create_annoset.py位置就行
   --anno-type=$anno_type 
   --label-map-file=$mapfile 
   --min-dim=$min_dim 
   --max-dim=$max_dim 
   --resize-width=$width 
   --resize-height=$height 
   # 很多人最后训练代码出错其实就是最后一行没有设置正确
   --check-label 
   $extra_cmd #这个不用管
   $data_root_dir #这个不用管
   $root_dir/data/$dataset_name/$subset.txt #这个位置是上述生成的trainval.txt和test.txt文件路径,这个要设置好
   $data_root_dir/$dataset_name/$db/$dataset_name"_"$subset"_"$db #这一个参数是设置生成lmdb文件的路径,一般来说,最好设置在VOC总文件夹下面,即同JPEGImages、Annotations、Images在同一个路径下,当然,也可以事先自己在VOC总文件夹下面创建一个lmdb的文件夹,该lmdb文件夹下面又有两个子文件夹,分别代表的是训练lmdb数据和测试lmdb数据
   examples/$dataset_name
done

注意点:
1.如果代码运行报错(明明图片和xml标签路径是正确的,但是就是无法生成lmdb),那么找到create_annoset.py第87行:

img_file, anno = line.strip("\n").split(" ")

改为

img_file, anno = line.strip("\n").strip("\r").split(" ")

2.生成LMDB文件,首先是sh create_data.sh,内部会调用python root_caffe_dir/scripts/create_annotation.py,但是真正调用的其实是root_caffe_dir/build/tools/convert_annoset的二进制文件,该文件又是由root_caffe_dir/tools/convert_annoset.cpp编译而来。打开root_caffe_dir/tools/convert_annoset.cpp,在大约157行左右可以看到:

filename = root_folder + lines[line_id].first;

所以说,官方代码默认带了caffe的根目录,所以如果想要匹配路径的话,就要修改你生成的trainval.txt和test.txt两个文件中图片的路径,或者是修改root_caffe_dir/tools/convert_annoset.cpp里面的代码,然后重新编译一边。

三、开始训练

与以往caffe实现分类网络不同的是,该项目是通过py脚本进行训练的,而不是直接通过caffe的c++接口进行训练。
该py脚本的位置在:you_caffe_root/examples/ssd/ssd_pascal.py。
跑通caffe-ssd demo代码(训练、测试自己数据集)_第11张图片
上述图片中还有一个ssd_detect.py脚本,该脚本就是用于测试单张图片用的,这个后续再说。
ssd_pascal.py
具体需要修改的地方会在下面标注清楚的

from __future__ import print_function
import caffe
from caffe.model_libs import *
from google.protobuf import text_format

import math
import os
import shutil
import stat
import subprocess
import sys

# Add extra layers on top of a "base" network (e.g. VGGNet or Inception).
def AddExtraLayers(net, use_batchnorm=True, lr_mult=1):
    use_relu = True

    # Add additional convolutional layers.
    # 19 x 19
    from_layer = net.keys()[-1]

    # TODO(weiliu89): Construct the name using the last layer to avoid duplication.
    # 10 x 10
    out_layer = "conv6_1"
    ConvBNLayer(net, from_layer, out_layer, use_batchnorm, use_relu, 256, 1, 0, 1,
        lr_mult=lr_mult)

    from_layer = out_layer
    out_layer = "conv6_2"
    ConvBNLayer(net, from_layer, out_layer, use_batchnorm, use_relu, 512, 3, 1, 2,
        lr_mult=lr_mult)

    # 5 x 5
    from_layer = out_layer
    out_layer = "conv7_1"
    ConvBNLayer(net, from_layer, out_layer, use_batchnorm, use_relu, 128, 1, 0, 1,
      lr_mult=lr_mult)

    from_layer = out_layer
    out_layer = "conv7_2"
    ConvBNLayer(net, from_layer, out_layer, use_batchnorm, use_relu, 256, 3, 1, 2,
      lr_mult=lr_mult)

    # 3 x 3
    from_layer = out_layer
    out_layer = "conv8_1"
    ConvBNLayer(net, from_layer, out_layer, use_batchnorm, use_relu, 128, 1, 0, 1,
      lr_mult=lr_mult)

    from_layer = out_layer
    out_layer = "conv8_2"
    ConvBNLayer(net, from_layer, out_layer, use_batchnorm, use_relu, 256, 3, 0, 1,
      lr_mult=lr_mult)

    # 1 x 1
    from_layer = out_layer
    out_layer = "conv9_1"
    ConvBNLayer(net, from_layer, out_layer, use_batchnorm, use_relu, 128, 1, 0, 1,
      lr_mult=lr_mult)

    from_layer = out_layer
    out_layer = "conv9_2"
    ConvBNLayer(net, from_layer, out_layer, use_batchnorm, use_relu, 256, 3, 0, 1,
      lr_mult=lr_mult)

    return net


### Modify the following parameters accordingly ###
# The directory which contains the caffe code.
# We assume you are running the script at the CAFFE_ROOT.
caffe_root = os.getcwd()

# Set true if you want to start training right after generating all files.
run_soon = True
# Set true if you want to load from most recently saved snapshot.
# Otherwise, we will load from the pretrain_model defined below.
resume_training = True
# If true, Remove old model files.
remove_old_models = False

# The database file for training data. Created by data/VOC0712/create_data.sh
train_data = "examples/VOC0712/VOC0712_trainval_lmdb"  #该地址为上述生成的lmdb格式的训练数据,注意的是,该地址为存放lmdb数据的上一级文件夹
# The database file for testing data. Created by data/VOC0712/create_data.sh
test_data = "examples/VOC0712/VOC0712_test_lmdb" #该地址为上述生成的lmdb格式的测试数据,注意的是,该地址为存放lmdb数据的上一级文件夹
# Specify the batch sampler.
resize_width = 300 #该模型接收输入图片的宽
resize_height = 300 #该模型接收输入图片的高
resize = "{}x{}".format(resize_width, resize_height)
batch_sampler = [
        {
                'sampler': {
                        },
                'max_trials': 1,
                'max_sample': 1,
        },
        {
                'sampler': {
                        'min_scale': 0.3,
                        'max_scale': 1.0,
                        'min_aspect_ratio': 0.5,
                        'max_aspect_ratio': 2.0,
                        },
                'sample_constraint': {
                        'min_jaccard_overlap': 0.1,
                        },
                'max_trials': 50,
                'max_sample': 1,
        },
        {
                'sampler': {
                        'min_scale': 0.3,
                        'max_scale': 1.0,
                        'min_aspect_ratio': 0.5,
                        'max_aspect_ratio': 2.0,
                        },
                'sample_constraint': {
                        'min_jaccard_overlap': 0.3,
                        },
                'max_trials': 50,
                'max_sample': 1,
        },
        {
                'sampler': {
                        'min_scale': 0.3,
                        'max_scale': 1.0,
                        'min_aspect_ratio': 0.5,
                        'max_aspect_ratio': 2.0,
                        },
                'sample_constraint': {
                        'min_jaccard_overlap': 0.5,
                        },
                'max_trials': 50,
                'max_sample': 1,
        },
        {
                'sampler': {
                        'min_scale': 0.3,
                        'max_scale': 1.0,
                        'min_aspect_ratio': 0.5,
                        'max_aspect_ratio': 2.0,
                        },
                'sample_constraint': {
                        'min_jaccard_overlap': 0.7,
                        },
                'max_trials': 50,
                'max_sample': 1,
        },
        {
                'sampler': {
                        'min_scale': 0.3,
                        'max_scale': 1.0,
                        'min_aspect_ratio': 0.5,
                        'max_aspect_ratio': 2.0,
                        },
                'sample_constraint': {
                        'min_jaccard_overlap': 0.9,
                        },
                'max_trials': 50,
                'max_sample': 1,
        },
        {
                'sampler': {
                        'min_scale': 0.3,
                        'max_scale': 1.0,
                        'min_aspect_ratio': 0.5,
                        'max_aspect_ratio': 2.0,
                        },
                'sample_constraint': {
                        'max_jaccard_overlap': 1.0,
                        },
                'max_trials': 50,
                'max_sample': 1,
        },
        ]
train_transform_param = {  #这个字典里面就是进行数据增强的操作,其实我感觉可以不要该操作,但是没有尝试过注释后代码还能不能跑通
        'mirror': True,
        'mean_value': [104, 117, 123], #图片均值
        'resize_param': {
                'prob': 1,
                'resize_mode': P.Resize.WARP,
                'height': resize_height,
                'width': resize_width,
                'interp_mode': [
                        P.Resize.LINEAR,
                        P.Resize.AREA,
                        P.Resize.NEAREST,
                        P.Resize.CUBIC,
                        P.Resize.LANCZOS4,
                        ],
                },
        'distort_param': {
                'brightness_prob': 0.5,
                'brightness_delta': 32,
                'contrast_prob': 0.5,
                'contrast_lower': 0.5,
                'contrast_upper': 1.5,
                'hue_prob': 0.5,
                'hue_delta': 18,
                'saturation_prob': 0.5,
                'saturation_lower': 0.5,
                'saturation_upper': 1.5,
                'random_order_prob': 0.0,
                },
        'expand_param': {
                'prob': 0.5,
                'max_expand_ratio': 4.0,
                },
        'emit_constraint': {
            'emit_type': caffe_pb2.EmitConstraint.CENTER,
            }
        }
test_transform_param = {
        'mean_value': [104, 117, 123],
        'resize_param': {
                'prob': 1,
                'resize_mode': P.Resize.WARP,
                'height': resize_height,
                'width': resize_width,
                'interp_mode': [P.Resize.LINEAR],
                },
        }

# If true, use batch norm for all newly added layers.
# Currently only the non batch norm version has been tested.
use_batchnorm = False
lr_mult = 1
# Use different initial learning rate.
if use_batchnorm:
    base_lr = 0.0004
else:
    # A learning rate for batch_size = 1, num_gpus = 1.
    base_lr = 0.00004  #一般来说,我们会调用这个学习率,但是实际在训练的时候的学习率应该为base_lr * 25,所以说如果想增减学习率,只需要修改此处就可以

# Modify the job name if you want.
job_name = "SSD_{}".format(resize)
# The name of the model. Modify it if you want.
model_name = "VGG_VOC0712_{}".format(job_name)

# Directory which stores the model .prototxt file.
save_dir = "models/VGGNet/VOC0712/{}".format(job_name) #最后生成caffemodel的位置
# Directory which stores the snapshot of models.
snapshot_dir = "models/VGGNet/VOC0712/{}".format(job_name) 
# Directory which stores the job script and log file.
job_dir = "jobs/VGGNet/VOC0712/{}".format(job_name)
# Directory which stores the detection results.
output_result_dir = "{}/data/VOCdevkit/results/VOC2007/{}/Main".format(os.environ['HOME'], job_name)

# model definition files.
train_net_file = "{}/train.prototxt".format(save_dir) #train.prototxt位置
test_net_file = "{}/test.prototxt".format(save_dir) #test.prototxt位置
deploy_net_file = "{}/deploy.prototxt".format(save_dir)
solver_file = "{}/solver.prototxt".format(save_dir)
# snapshot prefix.
snapshot_prefix = "{}/{}".format(snapshot_dir, model_name)
# job script path.
job_file = "{}/{}.sh".format(job_dir, model_name)

# Stores the test image names and sizes. Created by data/VOC0712/create_list.sh
name_size_file = "data/VOC0712/test_name_size.txt" #上述生成的test_name_size.txt位置,最后用绝对路径
# The pretrained model. We use the Fully convolutional reduced (atrous) VGGNet.
pretrain_model = "models/VGGNet/VGG_ILSVRC_16_layers_fc_reduced.caffemodel" #预训练模型,会很大程度减少自己项目的训练时间
# Stores LabelMapItem.
label_map_file = "data/VOC0712/labelmap_voc.prototxt" #上述生成的labelmap_voc.prototxt位置,最后用绝对路径

# MultiBoxLoss parameters.
num_classes = 21 #该位置要换成你自己项目的物体类别个数,别忘了要加上背景
share_location = True
background_label_id=0
train_on_diff_gt = True
normalization_mode = P.Loss.VALID
code_type = P.PriorBox.CENTER_SIZE
ignore_cross_boundary_bbox = False
mining_type = P.MultiBoxLoss.MAX_NEGATIVE
neg_pos_ratio = 3.
loc_weight = (neg_pos_ratio + 1.) / 4.
multibox_loss_param = {
    'loc_loss_type': P.MultiBoxLoss.SMOOTH_L1,
    'conf_loss_type': P.MultiBoxLoss.SOFTMAX,
    'loc_weight': loc_weight,
    'num_classes': num_classes,
    'share_location': share_location,
    'match_type': P.MultiBoxLoss.PER_PREDICTION,
    'overlap_threshold': 0.5,
    'use_prior_for_matching': True,
    'background_label_id': background_label_id,
    'use_difficult_gt': train_on_diff_gt,
    'mining_type': mining_type,
    'neg_pos_ratio': neg_pos_ratio,
    'neg_overlap': 0.5,
    'code_type': code_type,
    'ignore_cross_boundary_bbox': ignore_cross_boundary_bbox,
    }
loss_param = {
    'normalization': normalization_mode,
    }

# parameters for generating priors.
# minimum dimension of input image
min_dim = 300
# conv4_3 ==> 38 x 38
# fc7 ==> 19 x 19
# conv6_2 ==> 10 x 10
# conv7_2 ==> 5 x 5
# conv8_2 ==> 3 x 3
# conv9_2 ==> 1 x 1
mbox_source_layers = ['conv4_3', 'fc7', 'conv6_2', 'conv7_2', 'conv8_2', 'conv9_2']
# in percent %
min_ratio = 20
max_ratio = 90
step = int(math.floor((max_ratio - min_ratio) / (len(mbox_source_layers) - 2)))
min_sizes = []
max_sizes = []
for ratio in xrange(min_ratio, max_ratio + 1, step):
  min_sizes.append(min_dim * ratio / 100.)
  max_sizes.append(min_dim * (ratio + step) / 100.)
min_sizes = [min_dim * 10 / 100.] + min_sizes
max_sizes = [min_dim * 20 / 100.] + max_sizes
steps = [8, 16, 32, 64, 100, 300]
aspect_ratios = [[2], [2, 3], [2, 3], [2, 3], [2], [2]]
# L2 normalize conv4_3.
normalizations = [20, -1, -1, -1, -1, -1]
# variance used to encode/decode prior bboxes.
if code_type == P.PriorBox.CENTER_SIZE:
  prior_variance = [0.1, 0.1, 0.2, 0.2]
else:
  prior_variance = [0.1]
flip = True
clip = False

# Solver parameters.
# Defining which GPUs to use.
gpus = "0,1,2,3"
gpulist = gpus.split(",")
num_gpus = len(gpulist)

# Divide the mini-batch to different GPUs.
batch_size = 32 #批次训练大小,基于你自己机器的性能调节
accum_batch_size = 32
iter_size = accum_batch_size / batch_size
solver_mode = P.Solver.CPU
device_id = 0 #你的机器有几个GPU,这里就设置几
batch_size_per_device = batch_size
if num_gpus > 0:
  batch_size_per_device = int(math.ceil(float(batch_size) / num_gpus))
  iter_size = int(math.ceil(float(accum_batch_size) / (batch_size_per_device * num_gpus)))
  solver_mode = P.Solver.GPU
  device_id = int(gpulist[0])

if normalization_mode == P.Loss.NONE:
  base_lr /= batch_size_per_device
elif normalization_mode == P.Loss.VALID:
  base_lr *= 25. / loc_weight
elif normalization_mode == P.Loss.FULL:
  # Roughly there are 2000 prior bboxes per image.
  # TODO(weiliu89): Estimate the exact # of priors.
  base_lr *= 2000.

# Evaluate on whole test set.
num_test_image = 4952 #这个数字要与上述生成的test_name_size.txt的行数保持一致,及测试集一共多少张图片
test_batch_size = 8
# Ideally test_batch_size should be divisible by num_test_image,
# otherwise mAP will be slightly off the true value.
test_iter = int(math.ceil(float(num_test_image) / test_batch_size))

solver_param = {
    # Train parameters
    'base_lr': base_lr,
    'weight_decay': 0.0005,
    'lr_policy': "multistep",
    'stepvalue': [80000, 100000, 120000],
    'gamma': 0.1,
    'momentum': 0.9,
    'iter_size': iter_size,
    'max_iter': 120000,
    'snapshot': 80000,
    'display': 10,
    'average_loss': 10,
    'type': "SGD",
    'solver_mode': solver_mode,
    'device_id': device_id,
    'debug_info': False,
    'snapshot_after_train': True,
    # Test parameters
    'test_iter': [test_iter],
    'test_interval': 10000,
    'eval_type': "detection",
    'ap_version': "11point",
    'test_initialization': False,
    }

# parameters for generating detection output.
det_out_param = {
    'num_classes': num_classes,
    'share_location': share_location,
    'background_label_id': background_label_id,
    'nms_param': {'nms_threshold': 0.45, 'top_k': 400},
    'save_output_param': {
        'output_directory': output_result_dir,
        'output_name_prefix': "comp4_det_test_",
        'output_format': "VOC",
        'label_map_file': label_map_file,
        'name_size_file': name_size_file,
        'num_test_image': num_test_image,
        },
    'keep_top_k': 200,
    'confidence_threshold': 0.01,
    'code_type': code_type,
    }

# parameters for evaluating detection results.
det_eval_param = {
    'num_classes': num_classes,
    'background_label_id': background_label_id,
    'overlap_threshold': 0.5,
    'evaluate_difficult_gt': False,
    'name_size_file': name_size_file,
    }

### Hopefully you don't need to change the following ###
# Check file.
check_if_exist(train_data)
check_if_exist(test_data)
check_if_exist(label_map_file)
check_if_exist(pretrain_model)
make_if_not_exist(save_dir)
make_if_not_exist(job_dir)
make_if_not_exist(snapshot_dir)

# Create train net.
net = caffe.NetSpec()
net.data, net.label = CreateAnnotatedDataLayer(train_data, batch_size=batch_size_per_device,
        train=True, output_label=True, label_map_file=label_map_file,
        transform_param=train_transform_param, batch_sampler=batch_sampler)

VGGNetBody(net, from_layer='data', fully_conv=True, reduced=True, dilated=True,
    dropout=False)

AddExtraLayers(net, use_batchnorm, lr_mult=lr_mult)

mbox_layers = CreateMultiBoxHead(net, data_layer='data', from_layers=mbox_source_layers,
        use_batchnorm=use_batchnorm, min_sizes=min_sizes, max_sizes=max_sizes,
        aspect_ratios=aspect_ratios, steps=steps, normalizations=normalizations,
        num_classes=num_classes, share_location=share_location, flip=flip, clip=clip,
        prior_variance=prior_variance, kernel_size=3, pad=1, lr_mult=lr_mult)

# Create the MultiBoxLossLayer.
name = "mbox_loss"
mbox_layers.append(net.label)
net[name] = L.MultiBoxLoss(*mbox_layers, multibox_loss_param=multibox_loss_param,
        loss_param=loss_param, include=dict(phase=caffe_pb2.Phase.Value('TRAIN')),
        propagate_down=[True, True, False, False])

with open(train_net_file, 'w') as f:
    print('name: "{}_train"'.format(model_name), file=f)
    print(net.to_proto(), file=f)
shutil.copy(train_net_file, job_dir)

# Create test net.
net = caffe.NetSpec()
net.data, net.label = CreateAnnotatedDataLayer(test_data, batch_size=test_batch_size,
        train=False, output_label=True, label_map_file=label_map_file,
        transform_param=test_transform_param)

VGGNetBody(net, from_layer='data', fully_conv=True, reduced=True, dilated=True,
    dropout=False)

AddExtraLayers(net, use_batchnorm, lr_mult=lr_mult)

mbox_layers = CreateMultiBoxHead(net, data_layer='data', from_layers=mbox_source_layers,
        use_batchnorm=use_batchnorm, min_sizes=min_sizes, max_sizes=max_sizes,
        aspect_ratios=aspect_ratios, steps=steps, normalizations=normalizations,
        num_classes=num_classes, share_location=share_location, flip=flip, clip=clip,
        prior_variance=prior_variance, kernel_size=3, pad=1, lr_mult=lr_mult)

conf_name = "mbox_conf"
if multibox_loss_param["conf_loss_type"] == P.MultiBoxLoss.SOFTMAX:
  reshape_name = "{}_reshape".format(conf_name)
  net[reshape_name] = L.Reshape(net[conf_name], shape=dict(dim=[0, -1, num_classes]))
  softmax_name = "{}_softmax".format(conf_name)
  net[softmax_name] = L.Softmax(net[reshape_name], axis=2)
  flatten_name = "{}_flatten".format(conf_name)
  net[flatten_name] = L.Flatten(net[softmax_name], axis=1)
  mbox_layers[1] = net[flatten_name]
elif multibox_loss_param["conf_loss_type"] == P.MultiBoxLoss.LOGISTIC:
  sigmoid_name = "{}_sigmoid".format(conf_name)
  net[sigmoid_name] = L.Sigmoid(net[conf_name])
  mbox_layers[1] = net[sigmoid_name]

net.detection_out = L.DetectionOutput(*mbox_layers,
    detection_output_param=det_out_param,
    include=dict(phase=caffe_pb2.Phase.Value('TEST')))
net.detection_eval = L.DetectionEvaluate(net.detection_out, net.label,
    detection_evaluate_param=det_eval_param,
    include=dict(phase=caffe_pb2.Phase.Value('TEST')))

with open(test_net_file, 'w') as f:
    print('name: "{}_test"'.format(model_name), file=f)
    print(net.to_proto(), file=f)
shutil.copy(test_net_file, job_dir)

# Create deploy net.
# Remove the first and last layer from test net.
deploy_net = net
with open(deploy_net_file, 'w') as f:
    net_param = deploy_net.to_proto()
    # Remove the first (AnnotatedData) and last (DetectionEvaluate) layer from test net.
    del net_param.layer[0]
    del net_param.layer[-1]
    net_param.name = '{}_deploy'.format(model_name)
    net_param.input.extend(['data'])
    net_param.input_shape.extend([
        caffe_pb2.BlobShape(dim=[1, 3, resize_height, resize_width])])
    print(net_param, file=f)
shutil.copy(deploy_net_file, job_dir)

# Create solver.
solver = caffe_pb2.SolverParameter(
        train_net=train_net_file,
        test_net=[test_net_file],
        snapshot_prefix=snapshot_prefix,
        **solver_param)

with open(solver_file, 'w') as f:
    print(solver, file=f)
shutil.copy(solver_file, job_dir)

max_iter = 0
# Find most recent snapshot.
for file in os.listdir(snapshot_dir):
  if file.endswith(".solverstate"):
    basename = os.path.splitext(file)[0]
    iter = int(basename.split("{}_iter_".format(model_name))[1])
    if iter > max_iter:
      max_iter = iter

train_src_param = '--weights="{}" \\\n'.format(pretrain_model)
if resume_training:
  if max_iter > 0:
    train_src_param = '--snapshot="{}_iter_{}.solverstate" \\\n'.format(snapshot_prefix, max_iter)

if remove_old_models:
  # Remove any snapshots smaller than max_iter.
  for file in os.listdir(snapshot_dir):
    if file.endswith(".solverstate"):
      basename = os.path.splitext(file)[0]
      iter = int(basename.split("{}_iter_".format(model_name))[1])
      if max_iter > iter:
        os.remove("{}/{}".format(snapshot_dir, file))
    if file.endswith(".caffemodel"):
      basename = os.path.splitext(file)[0]
      iter = int(basename.split("{}_iter_".format(model_name))[1])
      if max_iter > iter:
        os.remove("{}/{}".format(snapshot_dir, file))

# Create job file.
with open(job_file, 'w') as f:
  f.write('cd {}\n'.format(caffe_root))
  f.write('./build/tools/caffe train \\\n')
  f.write('--solver="{}" \\\n'.format(solver_file))
  f.write(train_src_param)
  if solver_param['solver_mode'] == P.Solver.GPU:
    f.write('--gpu {} 2>&1 | tee {}/{}.log\n'.format(gpus, job_dir, model_name))
  else:
    f.write('2>&1 | tee {}/{}.log\n'.format(job_dir, model_name))

# Copy the python script to job_dir.
py_file = os.path.abspath(__file__)
shutil.copy(py_file, job_dir)

# Run the job.
os.chmod(job_file, stat.S_IRWXU)
if run_soon:
  subprocess.call(job_file, shell=True)

预训练模型文件下载地址:VGG_ILSVRC_16_layers_fc_reduced.caffemodel
注意点:
在脚本中能用绝对路径就用绝对路径,不然各种相对路径能让你心态爆炸!

四、测试模型

找到you_caffe_root/example/ssd/ssd_detect.py

#encoding=utf8
'''
Detection with SSD
In this example, we will load a SSD model and use it to detect objects.
'''

import os
import sys
import argparse
import numpy as np
from PIL import Image, ImageDraw
# Make sure that caffe is on the python path:
caffe_root = './'
os.chdir(caffe_root)
sys.path.insert(0, os.path.join(caffe_root, 'python'))
import caffe

from google.protobuf import text_format
from caffe.proto import caffe_pb2


def get_labelname(labelmap, labels):
    num_labels = len(labelmap.item)
    labelnames = []
    if type(labels) is not list:
        labels = [labels]
    for label in labels:
        found = False
        for i in xrange(0, num_labels):
            if label == labelmap.item[i].label:
                found = True
                labelnames.append(labelmap.item[i].display_name)
                break
        assert found == True
    return labelnames

class CaffeDetection:
    def __init__(self, gpu_id, model_def, model_weights, image_resize, labelmap_file):
        caffe.set_device(gpu_id)
        caffe.set_mode_gpu()

        self.image_resize = image_resize
        # Load the net in the test phase for inference, and configure input preprocessing.
        self.net = caffe.Net(model_def,      # defines the structure of the model
                             model_weights,  # contains the trained weights
                             caffe.TEST)     # use test mode (e.g., don't perform dropout)
         # input preprocessing: 'data' is the name of the input blob == net.inputs[0]
        self.transformer = caffe.io.Transformer({'data': self.net.blobs['data'].data.shape})
        self.transformer.set_transpose('data', (2, 0, 1))
        self.transformer.set_mean('data', np.array([104, 117, 123])) # mean pixel
        # the reference model operates on images in [0,255] range instead of [0,1]
        self.transformer.set_raw_scale('data', 255)
        # the reference model has channels in BGR order instead of RGB
        self.transformer.set_channel_swap('data', (2, 1, 0))

        # load PASCAL VOC labels
        file = open(labelmap_file, 'r')
        self.labelmap = caffe_pb2.LabelMap()
        text_format.Merge(str(file.read()), self.labelmap)

    def detect(self, image_file, conf_thresh=0.5, topn=5):
        '''
        SSD detection
        '''
        # set net to batch size of 1
        # image_resize = 300
        self.net.blobs['data'].reshape(1, 3, self.image_resize, self.image_resize)
        image = caffe.io.load_image(image_file)

        #Run the net and examine the top_k results
        transformed_image = self.transformer.preprocess('data', image)
        self.net.blobs['data'].data[...] = transformed_image

        # Forward pass.
        detections = self.net.forward()['detection_out']

        # Parse the outputs.
        det_label = detections[0,0,:,1]
        det_conf = detections[0,0,:,2]
        det_xmin = detections[0,0,:,3]
        det_ymin = detections[0,0,:,4]
        det_xmax = detections[0,0,:,5]
        det_ymax = detections[0,0,:,6]

        # Get detections with confidence higher than 0.6.
        top_indices = [i for i, conf in enumerate(det_conf) if conf >= conf_thresh]

        top_conf = det_conf[top_indices]
        top_label_indices = det_label[top_indices].tolist()
        top_labels = get_labelname(self.labelmap, top_label_indices)
        top_xmin = det_xmin[top_indices]
        top_ymin = det_ymin[top_indices]
        top_xmax = det_xmax[top_indices]
        top_ymax = det_ymax[top_indices]

        result = []
        for i in xrange(min(topn, top_conf.shape[0])):
            xmin = top_xmin[i] # xmin = int(round(top_xmin[i] * image.shape[1]))
            ymin = top_ymin[i] # ymin = int(round(top_ymin[i] * image.shape[0]))
            xmax = top_xmax[i] # xmax = int(round(top_xmax[i] * image.shape[1]))
            ymax = top_ymax[i] # ymax = int(round(top_ymax[i] * image.shape[0]))
            score = top_conf[i]
            label = int(top_label_indices[i])
            label_name = top_labels[i]
            result.append([xmin, ymin, xmax, ymax, label, score, label_name])
        return result

def main(args):
    '''main '''
    detection = CaffeDetection(args.gpu_id,
                               args.model_def, args.model_weights,
                               args.image_resize, args.labelmap_file)
    result = detection.detect(args.image_file)
    print result

    img = Image.open(args.image_file)
    draw = ImageDraw.Draw(img)
    width, height = img.size
    print width, height
    for item in result:
        xmin = int(round(item[0] * width))
        ymin = int(round(item[1] * height))
        xmax = int(round(item[2] * width))
        ymax = int(round(item[3] * height))
        draw.rectangle([xmin, ymin, xmax, ymax], outline=(255, 0, 0))
        draw.text([xmin, ymin], item[-1] + str(item[-2]), (0, 0, 255))
        print item
        print [xmin, ymin, xmax, ymax]
        print [xmin, ymin], item[-1]
    img.save('detect_result.jpg')


def parse_args():
    '''parse args'''
    parser = argparse.ArgumentParser()
    parser.add_argument('--gpu_id', type=int, default=0, help='gpu id')
    parser.add_argument('--labelmap_file',
                        default='data/VOC0712/labelmap_voc.prototxt') #上述生成的labelmap_voc.prototxt位置
    parser.add_argument('--model_def',
                        default='models/VGGNet/VOC0712/SSD_300x300/deploy.prototxt') #训练阶段生成的测试prototxt文件路径
    parser.add_argument('--image_resize', default=300, type=int) #测试图片大小要与训练一致
    parser.add_argument('--model_weights',
                        default='models/VGGNet/VOC0712/SSD_300x300/'
                        'VGG_VOC0712_SSD_300x300_iter_120000.caffemodel') #训练阶段生成的caffemodel路径
    parser.add_argument('--image_file', default='examples/images/fish-bike.jpg') #要测试图片的路径
    return parser.parse_args()

if __name__ == '__main__':
    main(parse_args())

https://blog.csdn.net/chenlufei_i/article/details/80068953
https://blog.csdn.net/weixin_30613343/article/details/96541315

你可能感兴趣的:(目标检测,Caffe,SSD,python,计算机视觉,人工智能,caffe,深度学习)