main.cpp 11.2 KB
Newer Older
1

2
#include "dlib/data_io.h"
3
#include "dlib/string.h"
4
#include "metadata_editor.h"
Davis King's avatar
Davis King committed
5
#include "convert_pascal_xml.h"
6
#include "convert_pascal_v1.h"
7
#include "convert_idl.h"
Davis King's avatar
Davis King committed
8
#include <dlib/cmd_line_parser.h>
9

10
11
#include <iostream>
#include <fstream>
12
#include <string>
13
#include <set>
14

15
#include <dlib/dir_nav.h>
16
17


18
const char* VERSION = "0.5";
Davis King's avatar
Davis King committed
19
20


21
22
23
24

using namespace std;
using namespace dlib;

25
26
// ----------------------------------------------------------------------------------------

Davis King's avatar
Davis King committed
27
void create_new_dataset (
Davis King's avatar
Davis King committed
28
    const command_line_parser& parser
Davis King's avatar
Davis King committed
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
)
{
    using namespace dlib::image_dataset_metadata;

    const std::string filename = parser.option("c").argument();
    // make sure the file exists so we can use the get_parent_directory() command to
    // figure out it's parent directory.
    make_empty_file(filename);
    const std::string parent_dir = get_parent_directory(file(filename)).full_name();

    unsigned long depth = 0;
    if (parser.option("r"))
        depth = 30;

    dataset meta;
    meta.name = "imglab dataset";
    meta.comment = "Created by imglab tool.";
    for (unsigned long i = 0; i < parser.number_of_arguments(); ++i)
    {
        try
        {
            const string temp = strip_path(file(parser[i]).full_name(), parent_dir);
            meta.images.push_back(image(temp));
        }
        catch (dlib::file::file_not_found&)
        {
            // then parser[i] should be a directory

            std::vector<file> files = get_files_in_directory_tree(parser[i], 
                                                                  match_endings(".png .PNG .jpeg .JPEG .jpg .JPG .bmp .BMP .dng .DNG"),
                                                                  depth);
            sort(files.begin(), files.end());

            for (unsigned long j = 0; j < files.size(); ++j)
            {
                meta.images.push_back(image(strip_path(files[j].full_name(), parent_dir)));
            }
        }
    }

    save_image_dataset_metadata(meta, filename);
}

// ----------------------------------------------------------------------------------------
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
void print_all_labels (
    const dlib::image_dataset_metadata::dataset& data
)
{
    std::set<std::string> labels;
    for (unsigned long i = 0; i < data.images.size(); ++i)
    {
        for (unsigned long j = 0; j < data.images[i].boxes.size(); ++j)
        {
            labels.insert(data.images[i].boxes[j].label);
        }
    }

    for (std::set<std::string>::iterator i = labels.begin(); i != labels.end(); ++i)
    {
        if (i->size() != 0)
        {
            cout << *i << endl;
        }
    }
}

// ----------------------------------------------------------------------------------------

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
void print_all_label_stats (
    const dlib::image_dataset_metadata::dataset& data
)
{
    std::map<std::string, running_stats<double> > area_stats, aspect_ratio;
    std::set<std::string> labels;
    for (unsigned long i = 0; i < data.images.size(); ++i)
    {
        for (unsigned long j = 0; j < data.images[i].boxes.size(); ++j)
        {
            labels.insert(data.images[i].boxes[j].label);

            area_stats[data.images[i].boxes[j].label].add(data.images[i].boxes[j].rect.area());
            aspect_ratio[data.images[i].boxes[j].label].add(data.images[i].boxes[j].rect.width()/
                                                    (double)data.images[i].boxes[j].rect.height());
        }
    }

    cout << "Number of different labels: "<< labels.size() << endl << endl;

    for (std::set<std::string>::iterator i = labels.begin(); i != labels.end(); ++i)
    {
        if (i->size() != 0)
        {
            cout << "Label: "<< *i << endl;
            cout << "   number of occurrences: " << area_stats[*i].current_n() << endl;
            cout << "   min box area:    " << area_stats[*i].min() << endl;
            cout << "   max box area:    " << area_stats[*i].max() << endl;
            cout << "   mean box area:   " << area_stats[*i].mean() << endl;
            cout << "   stddev box area: " << area_stats[*i].stddev() << endl;
            cout << "   mean width/height ratio:   " << aspect_ratio[*i].mean() << endl;
            cout << "   stddev width/height ratio: " << aspect_ratio[*i].stddev() << endl;
            cout << endl;
        }
    }
}

// ----------------------------------------------------------------------------------------

Davis King's avatar
Davis King committed
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
void rename_labels (
    dlib::image_dataset_metadata::dataset& data,
    const std::string& from,
    const std::string& to
)
{
    for (unsigned long i = 0; i < data.images.size(); ++i)
    {
        for (unsigned long j = 0; j < data.images[i].boxes.size(); ++j)
        {
            if (data.images[i].boxes[j].label == from)
                data.images[i].boxes[j].label = to;
        }
    }

}

// ----------------------------------------------------------------------------------------

156
157
158
159
160
int main(int argc, char** argv)
{
    try
    {

Davis King's avatar
Davis King committed
161
        command_line_parser parser;
162
163

        parser.add_option("h","Displays this information.");
164
165
166
        parser.add_option("v","Display version.");

        parser.set_group_name("Creating XML files");
167
        parser.add_option("c","Create an XML file named <arg> listing a set of images.",1);
168
        parser.add_option("r","Search directories recursively for images.");
169
170
171
172
        parser.add_option("convert","Convert foreign image Annotations from <arg> format to the imglab format. "
                          "Supported formats: pascal-xml, pascal-v1, idl.",1);

        parser.set_group_name("Viewing/Editing XML files");
173
        parser.add_option("l","List all the labels in the given XML file.");
174
        parser.add_option("stats","List detailed statistics on the object labels in the given XML file.");
Davis King's avatar
Davis King committed
175
        parser.add_option("rename", "Rename all labels of <arg1> to <arg2>.",2);
176
177
        parser.add_option("parts","The display will allow image parts to be labeled.  The set of allowable parts "
                          "defined in a space separated list contained in <arg>.",1);
178
        parser.add_option("rmdiff","Remove boxes marked as difficult.");
179
180
181

        parser.parse(argc, argv);

182
        const char* singles[] = {"h","c","r","l","convert","parts","rmdiff"};
183
        parser.check_one_time_options(singles);
184
185
        const char* c_sub_ops[] = {"r", "convert"};
        parser.check_sub_options("c", c_sub_ops);
Davis King's avatar
Davis King committed
186
        parser.check_incompatible_options("c", "l");
187
        parser.check_incompatible_options("c", "rmdiff");
Davis King's avatar
Davis King committed
188
        parser.check_incompatible_options("c", "rename");
189
        parser.check_incompatible_options("c", "parts");
Davis King's avatar
Davis King committed
190
        parser.check_incompatible_options("l", "rename");
191
        parser.check_incompatible_options("l", "parts");
192
193
        parser.check_incompatible_options("convert", "l");
        parser.check_incompatible_options("convert", "rename");
194
        parser.check_incompatible_options("convert", "parts");
195
        parser.check_incompatible_options("rmdiff", "rename");
196
        const char* convert_args[] = {"pascal-xml","pascal-v1","idl"};
197
        parser.check_option_arg_range("convert", convert_args);
198
199
200

        if (parser.option("h"))
        {
201
            cout << "Usage: imglab [options] <image files/directories or XML file>\n";
202
            parser.print_options(cout);
Davis King's avatar
Davis King committed
203
            cout << endl << endl;
204
205
206
            return EXIT_SUCCESS;
        }

Davis King's avatar
Davis King committed
207
208
209
210
211
212
213
214
215
        if (parser.option("v"))
        {
            cout << "imglab v" << VERSION 
                 << "\nCompiled: " << __TIME__ << " " << __DATE__ 
                 << "\nWritten by Davis King\n";
            cout << "Check for updates at http://dlib.net\n\n";
            return EXIT_SUCCESS;
        }

216
217
        if (parser.option("c"))
        {
218
219
            if (parser.option("convert"))
            {
Davis King's avatar
Davis King committed
220
221
                if (parser.option("convert").argument() == "pascal-xml")
                    convert_pascal_xml(parser);
222
223
                else if (parser.option("convert").argument() == "pascal-v1")
                    convert_pascal_v1(parser);
224
225
                else if (parser.option("convert").argument() == "idl")
                    convert_idl(parser);
226
227
228
229
230
            }
            else
            {
                create_new_dataset(parser);
            }
231
232
            return EXIT_SUCCESS;
        }
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
        
        if (parser.option("rmdiff"))
        {
            if (parser.number_of_arguments() != 1)
            {
                cerr << "The --rmdiff option requires you to give one XML file on the command line." << endl;
                return EXIT_FAILURE;
            }

            dlib::image_dataset_metadata::dataset data;
            load_image_dataset_metadata(data, parser[0]);
            for (unsigned long i = 0; i < data.images.size(); ++i)
            {
                std::vector<dlib::image_dataset_metadata::box> boxes;
                for (unsigned long j = 0; j < data.images[i].boxes.size(); ++j)
                {
                    if (!data.images[i].boxes[j].difficult)
                        boxes.push_back(data.images[i].boxes[j]);
                }
                data.images[i].boxes = boxes;
            }
            save_image_dataset_metadata(data, parser[0]);
            return EXIT_SUCCESS;
        }
257

258
259
260
261
262
263
264
265
266
267
268
269
270
271
        if (parser.option("l"))
        {
            if (parser.number_of_arguments() != 1)
            {
                cerr << "The -l option requires you to give one XML file on the command line." << endl;
                return EXIT_FAILURE;
            }

            dlib::image_dataset_metadata::dataset data;
            load_image_dataset_metadata(data, parser[0]);
            print_all_labels(data);
            return EXIT_SUCCESS;
        }

272
273
274
275
276
277
278
279
280
281
282
283
284
285
        if (parser.option("stats"))
        {
            if (parser.number_of_arguments() != 1)
            {
                cerr << "The --stats option requires you to give one XML file on the command line." << endl;
                return EXIT_FAILURE;
            }

            dlib::image_dataset_metadata::dataset data;
            load_image_dataset_metadata(data, parser[0]);
            print_all_label_stats(data);
            return EXIT_SUCCESS;
        }

Davis King's avatar
Davis King committed
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
        if (parser.option("rename"))
        {
            if (parser.number_of_arguments() != 1)
            {
                cerr << "The --rename option requires you to give one XML file on the command line." << endl;
                return EXIT_FAILURE;
            }

            dlib::image_dataset_metadata::dataset data;
            load_image_dataset_metadata(data, parser[0]);
            for (unsigned long i = 0; i < parser.option("rename").count(); ++i)
            {
                rename_labels(data, parser.option("rename").argument(0,i), parser.option("rename").argument(1,i));
            }
            save_image_dataset_metadata(data, parser[0]);
            return EXIT_SUCCESS;
        }

304
305
        if (parser.number_of_arguments() == 1)
        {
306
            metadata_editor editor(parser[0]);
307
308
309
310
311
312
313
314
            if (parser.option("parts"))
            {
                std::vector<string> parts = split(parser.option("parts").argument());
                for (unsigned long i = 0; i < parts.size(); ++i)
                {
                    editor.add_labelable_part_name(parts[i]);
                }
            }
315
            editor.wait_until_closed();
316
        }
317
318
319
    }
    catch (exception& e)
    {
320
        cerr << e.what() << endl;
321
322
323
324
        return EXIT_FAILURE;
    }
}

Davis King's avatar
Davis King committed
325
326
// ----------------------------------------------------------------------------------------