int main(int argc,
char **argv)
{
char *vers;
PIX *pixs;
l_int32 i;
l_int32 nWriteTries = REPETITIONS;
static char mainName[] = "gifio_leaktest";
vers = getLeptonicaVersion();
fprintf(stderr, "%s\n", vers);
FREE(vers);
vers = getImagelibVersions();
fprintf(stderr, "%s\n", vers);
FREE(vers);
if (argc > 2)
return ERROR_INT(" Syntax: gifio_leaktest [skipWriteTests]",
mainName, 1);
if (argc == 2)
nWriteTries = 0;
if ((pixs = pixRead("church.png")) == NULL)
return ERROR_INT("pix not found", mainName, 1);
for (i = 0; i < nWriteTries; i++) {
fprintf(stderr, "Writing attempt %d\n", i+1);
if (pixWrite("/tmp/church.gif", pixs, IFF_GIF)) {
fprintf(stderr, "Failed to write gif file on %dth try.\n", i+1);
break;
}
}
pixDestroy(&pixs);
if (i == REPETITIONS)
fprintf(stderr, "Successfully wrote gif file %d times.\n", i);
for (i = 0; i < REPETITIONS; i++) {
fprintf(stderr, "Reading attempt %d\n", i+1);
pixs = pixRead("/tmp/church.gif");
if (pixs == NULL) {
fprintf(stderr, "Failed to read gif file on %dth try.\n", i+1);
break;
} else {
pixDestroy(&pixs);
}
}
if (i == REPETITIONS)
fprintf(stderr, "Successfully read gif file %d times.\n", i);
return 0;
}
void PrintVersionInfo() {
char* versionStrP;
printf("tesseract %s\n", tesseract::TessBaseAPI::Version());
versionStrP = getLeptonicaVersion();
printf(" %s\n", versionStrP);
lept_free(versionStrP);
versionStrP = getImagelibVersions();
printf(" %s\n", versionStrP);
lept_free(versionStrP);
#ifdef USE_OPENCL
cl_platform_id platform[4];
cl_uint num_platforms;
printf(" OpenCL info:\n");
if (clGetPlatformIDs(4, platform, &num_platforms) == CL_SUCCESS) {
printf(" Found %u platform(s).\n", num_platforms);
for (unsigned n = 0; n < num_platforms; n++) {
char info[256];
if (clGetPlatformInfo(platform[n], CL_PLATFORM_NAME, 256, info, 0) == CL_SUCCESS) {
printf(" Platform %u name: %s.\n", n + 1, info);
}
if (clGetPlatformInfo(platform[n], CL_PLATFORM_VERSION, 256, info, 0) == CL_SUCCESS) {
printf(" Version: %s.\n", info);
}
cl_device_id devices[2];
cl_uint num_devices;
if (clGetDeviceIDs(platform[n], CL_DEVICE_TYPE_ALL, 2, devices, &num_devices) == CL_SUCCESS) {
printf(" Found %u device(s).\n", num_devices);
for (unsigned i = 0; i < num_devices; ++i) {
if (clGetDeviceInfo(devices[i], CL_DEVICE_NAME, 256, info, 0) == CL_SUCCESS) {
printf(" Device %u name: %s.\n", i + 1, info);
}
}
}
}
}
#endif
if (SIMDDetect::IsAVXAvailable())
printf(" Found AVX\n");
if (SIMDDetect::IsSSEAvailable())
printf(" Found SSE\n");
}
/*
* Provide information about application
*/
void MainWindow::about() {
QString abouttext =
tr("<h1>%1 %2</h1>").arg(SETTING_APPLICATION).arg(VERSION);
abouttext.append(tr("<p>playground for leptonica and "));
abouttext.append(tr("<a href=\"http://www.qt.io/\">Qt</a></p>"));
abouttext.append(tr("<p><b>Leptonica version:</b><br/> %1<br/>").arg(
getLeptonicaVersion()));
abouttext.append(tr("<b>Image libraries in Leptonica:</b><br/> %1</p>").arg(
getImagelibVersions()));
abouttext.append(tr("<p>Project page: <a href=%1>%2</a></p>").
arg(PROJECT_URL).arg(PROJECT_URL_NAME));
abouttext.append(tr("Copyright 2015 Zdenko Podobný</p>"));
abouttext.append(tr("<p>This software is released under "
"<a href=\"http://www.apache.org/licenses/LICENSE-2.0\"" \
">Apache License 2.0</a></p>"));
QMessageBox::about(this, tr("About application"), abouttext);
}
void PrintVersionInfo() {
char *versionStrP;
fprintf(stderr, "tesseract %s\n", tesseract::TessBaseAPI::Version());
versionStrP = getLeptonicaVersion();
fprintf(stderr, " %s\n", versionStrP);
lept_free(versionStrP);
versionStrP = getImagelibVersions();
fprintf(stderr, " %s\n", versionStrP);
lept_free(versionStrP);
#ifdef USE_OPENCL
cl_platform_id platform;
cl_uint num_platforms;
cl_device_id devices[2];
cl_uint num_devices;
char info[256];
int i;
fprintf(stderr, " OpenCL info:\n");
clGetPlatformIDs(1, &platform, &num_platforms);
fprintf(stderr, " Found %d platforms.\n", num_platforms);
clGetPlatformInfo(platform, CL_PLATFORM_NAME, 256, info, 0);
fprintf(stderr, " Platform name: %s.\n", info);
clGetPlatformInfo(platform, CL_PLATFORM_VERSION, 256, info, 0);
fprintf(stderr, " Version: %s.\n", info);
clGetDeviceIDs(platform, CL_DEVICE_TYPE_ALL, 2, devices, &num_devices);
fprintf(stderr, " Found %d devices.\n", num_devices);
for (i = 0; i < num_devices; ++i) {
clGetDeviceInfo(devices[i], CL_DEVICE_NAME, 256, info, 0);
fprintf(stderr, " Device %d name: %s.\n", i+1, info);
}
#endif
}
int main(int argc, char **argv) {
if ((argc == 2 && strcmp(argv[1], "-v") == 0) ||
(argc == 2 && strcmp(argv[1], "--version") == 0)) {
char *versionStrP;
fprintf(stderr, "tesseract %s\n", tesseract::TessBaseAPI::Version());
versionStrP = getLeptonicaVersion();
fprintf(stderr, " %s\n", versionStrP);
lept_free(versionStrP);
versionStrP = getImagelibVersions();
fprintf(stderr, " %s\n", versionStrP);
lept_free(versionStrP);
#ifdef USE_OPENCL
cl_platform_id platform;
cl_uint num_platforms;
cl_device_id devices[2];
cl_uint num_devices;
char info[256];
int i;
fprintf(stderr, " OpenCL info:\n");
clGetPlatformIDs(1, &platform, &num_platforms);
fprintf(stderr, " Found %d platforms.\n", num_platforms);
clGetPlatformInfo(platform, CL_PLATFORM_NAME, 256, info, 0);
fprintf(stderr, " Platform name: %s.\n", info);
clGetPlatformInfo(platform, CL_PLATFORM_VERSION, 256, info, 0);
fprintf(stderr, " Version: %s.\n", info);
clGetDeviceIDs(platform, CL_DEVICE_TYPE_ALL, 2, devices, &num_devices);
fprintf(stderr, " Found %d devices.\n", num_devices);
for (i = 0; i < num_devices; ++i) {
clGetDeviceInfo(devices[i], CL_DEVICE_NAME, 256, info, 0);
fprintf(stderr, " Device %d name: %s.\n", i+1, info);
}
#endif
exit(0);
}
// Make the order of args a bit more forgiving than it used to be.
const char* lang = "eng";
const char* image = NULL;
const char* outputbase = NULL;
const char* datapath = NULL;
bool noocr = false;
bool list_langs = false;
bool print_parameters = false;
GenericVector<STRING> vars_vec, vars_values;
tesseract::PageSegMode pagesegmode = tesseract::PSM_AUTO;
int arg = 1;
while (arg < argc && (outputbase == NULL || argv[arg][0] == '-')) {
if (strcmp(argv[arg], "-l") == 0 && arg + 1 < argc) {
lang = argv[arg + 1];
++arg;
} else if (strcmp(argv[arg], "--tessdata-dir") == 0 && arg + 1 < argc) {
datapath = argv[arg + 1];
++arg;
} else if (strcmp(argv[arg], "--user-words") == 0 && arg + 1 < argc) {
vars_vec.push_back("user_words_file");
vars_values.push_back(argv[arg + 1]);
++arg;
} else if (strcmp(argv[arg], "--user-patterns") == 0 && arg + 1 < argc) {
vars_vec.push_back("user_patterns_file");
vars_values.push_back(argv[arg + 1]);
++arg;
} else if (strcmp(argv[arg], "--list-langs") == 0) {
noocr = true;
list_langs = true;
} else if (strcmp(argv[arg], "-psm") == 0 && arg + 1 < argc) {
pagesegmode = static_cast<tesseract::PageSegMode>(atoi(argv[arg + 1]));
++arg;
} else if (strcmp(argv[arg], "--print-parameters") == 0) {
noocr = true;
print_parameters = true;
} else if (strcmp(argv[arg], "-c") == 0 && arg + 1 < argc) {
// handled properly after api init
++arg;
} else if (image == NULL) {
image = argv[arg];
} else if (outputbase == NULL) {
outputbase = argv[arg];
}
++arg;
}
if (argc == 2 && strcmp(argv[1], "--list-langs") == 0) {
list_langs = true;
noocr = true;
}
if (outputbase == NULL && noocr == false) {
fprintf(stderr, "Usage:\n %s imagename|stdin outputbase|stdout "
"[options...] [configfile...]\n\n", argv[0]);
fprintf(stderr, "OCR options:\n");
fprintf(stderr, " --tessdata-dir /path\tspecify the location of tessdata"
" path\n");
fprintf(stderr, " --user-words /path/to/file\tspecify the location of user"
" words file\n");
fprintf(stderr, " --user-patterns /path/to/file\tspecify the location of"
" user patterns file\n");
fprintf(stderr, " -l lang[+lang]\tspecify language(s) used for OCR\n");
fprintf(stderr, " -c configvar=value\tset value for control parameter.\n"
"\t\t\tMultiple -c arguments are allowed.\n");
fprintf(stderr, " -psm pagesegmode\tspecify page segmentation mode.\n");
fprintf(stderr, "These options must occur before any configfile.\n\n");
fprintf(stderr,
"pagesegmode values are:\n"
" 0 = Orientation and script detection (OSD) only.\n"
" 1 = Automatic page segmentation with OSD.\n"
" 2 = Automatic page segmentation, but no OSD, or OCR\n"
" 3 = Fully automatic page segmentation, but no OSD. (Default)\n"
" 4 = Assume a single column of text of variable sizes.\n"
" 5 = Assume a single uniform block of vertically aligned text.\n"
" 6 = Assume a single uniform block of text.\n"
" 7 = Treat the image as a single text line.\n"
" 8 = Treat the image as a single word.\n"
" 9 = Treat the image as a single word in a circle.\n"
" 10 = Treat the image as a single character.\n\n");
fprintf(stderr, "Single options:\n");
fprintf(stderr, " -v --version: version info\n");
fprintf(stderr, " --list-langs: list available languages for tesseract "
"engine. Can be used with --tessdata-dir.\n");
fprintf(stderr, " --print-parameters: print tesseract parameters to the "
"stdout.\n");
exit(1);
}
if (outputbase != NULL && strcmp(outputbase, "-") &&
strcmp(outputbase, "stdout")) {
tprintf("Tesseract Open Source OCR Engine v%s with Leptonica\n",
tesseract::TessBaseAPI::Version());
}
PERF_COUNT_START("Tesseract:main")
tesseract::TessBaseAPI api;
api.SetOutputName(outputbase);
int rc = api.Init(datapath, lang, tesseract::OEM_DEFAULT,
&(argv[arg]), argc - arg, &vars_vec, &vars_values, false);
if (rc) {
fprintf(stderr, "Could not initialize tesseract.\n");
exit(1);
}
char opt1[255], opt2[255];
for (arg = 0; arg < argc; arg++) {
if (strcmp(argv[arg], "-c") == 0 && arg + 1 < argc) {
strncpy(opt1, argv[arg + 1], 255);
char *p = strchr(opt1, '=');
if (!p) {
fprintf(stderr, "Missing = in configvar assignment\n");
exit(1);
}
*p = 0;
strncpy(opt2, strchr(argv[arg + 1], '=') + 1, 255);
opt2[254] = 0;
++arg;
if (!api.SetVariable(opt1, opt2)) {
fprintf(stderr, "Could not set option: %s=%s\n", opt1, opt2);
}
}
}
if (list_langs) {
GenericVector<STRING> languages;
api.GetAvailableLanguagesAsVector(&languages);
fprintf(stderr, "List of available languages (%d):\n",
languages.size());
for (int index = 0; index < languages.size(); ++index) {
STRING& string = languages[index];
fprintf(stderr, "%s\n", string.string());
}
api.End();
exit(0);
}
if (print_parameters) {
FILE* fout = stdout;
fprintf(stdout, "Tesseract parameters:\n");
api.PrintVariables(fout);
api.End();
exit(0);
}
// We have 2 possible sources of pagesegmode: a config file and
// the command line. For backwards compatibility reasons, the
// default in tesseract is tesseract::PSM_SINGLE_BLOCK, but the
// default for this program is tesseract::PSM_AUTO. We will let
// the config file take priority, so the command-line default
// can take priority over the tesseract default, so we use the
// value from the command line only if the retrieved mode
// is still tesseract::PSM_SINGLE_BLOCK, indicating no change
// in any config file. Therefore the only way to force
// tesseract::PSM_SINGLE_BLOCK is from the command line.
// It would be simpler if we could set the value before Init,
// but that doesn't work.
if (api.GetPageSegMode() == tesseract::PSM_SINGLE_BLOCK)
api.SetPageSegMode(pagesegmode);
if (pagesegmode == tesseract::PSM_AUTO_ONLY ||
pagesegmode == tesseract::PSM_OSD_ONLY) {
int ret_val = 0;
Pix* pixs = pixRead(image);
if (!pixs) {
fprintf(stderr, "Cannot open input file: %s\n", image);
exit(2);
}
api.SetImage(pixs);
if (pagesegmode == tesseract::PSM_OSD_ONLY) {
OSResults osr;
if (api.DetectOS(&osr)) {
int orient = osr.best_result.orientation_id;
int script_id = osr.get_best_script(orient);
float orient_oco = osr.best_result.oconfidence;
float orient_sco = osr.best_result.sconfidence;
tprintf("Orientation: %d\nOrientation in degrees: %d\n" \
"Orientation confidence: %.2f\n" \
"Script: %d\nScript confidence: %.2f\n",
orient, OrientationIdToValue(orient), orient_oco,
script_id, orient_sco);
} else {
ret_val = 1;
}
} else {
tesseract::Orientation orientation;
tesseract::WritingDirection direction;
tesseract::TextlineOrder order;
float deskew_angle;
tesseract::PageIterator* it = api.AnalyseLayout();
if (it) {
it->Orientation(&orientation, &direction, &order, &deskew_angle);
tprintf("Orientation: %d\nWritingDirection: %d\nTextlineOrder: %d\n" \
"Deskew angle: %.4f\n",
orientation, direction, order, deskew_angle);
} else {
ret_val = 1;
}
delete it;
}
pixDestroy(&pixs);
exit(ret_val);
}
bool b;
tesseract::PointerVector<tesseract::TessResultRenderer> renderers;
api.GetBoolVariable("tessedit_create_hocr", &b);
if (b) {
bool font_info;
api.GetBoolVariable("hocr_font_info", &font_info);
renderers.push_back(new tesseract::TessHOcrRenderer(outputbase, font_info));
}
api.GetBoolVariable("tessedit_create_pdf", &b);
if (b) {
renderers.push_back(new tesseract::TessPDFRenderer(outputbase,
api.GetDatapath()));
}
api.GetBoolVariable("tessedit_write_unlv", &b);
if (b) renderers.push_back(new tesseract::TessUnlvRenderer(outputbase));
api.GetBoolVariable("tessedit_create_boxfile", &b);
if (b) renderers.push_back(new tesseract::TessBoxTextRenderer(outputbase));
api.GetBoolVariable("tessedit_create_txt", &b);
if (b) renderers.push_back(new tesseract::TessTextRenderer(outputbase));
if (!renderers.empty()) {
// Since the PointerVector auto-deletes, null-out the renderers that are
// added to the root, and leave the root in the vector.
for (int r = 1; r < renderers.size(); ++r) {
renderers[0]->insert(renderers[r]);
renderers[r] = NULL;
}
if (!api.ProcessPages(image, NULL, 0, renderers[0])) {
fprintf(stderr, "Error during processing.\n");
exit(1);
}
}
PERF_COUNT_END
return 0; // Normal exit
}
/*!
* regTestSetup()
*
* Input: argc (from invocation; can be either 1 or 2)
* argv (to regtest: @argv[1] is one of these:
* "generate", "compare", "display")
* &rp (<return> all regression params)
* Return: 0 if OK, 1 on error
*
* Notes:
* (1) Call this function with the args to the reg test.
* There are three cases:
* Case 1:
* There is either only one arg, or the second arg is "compare".
* This is the mode in which you run a regression test
* (or a set of them), looking for failures and logging
* the results to a file. The output, which includes
* logging of all reg test failures plus a SUCCESS or
* FAILURE summary for each test, is appended to the file
* "/tmp/reg_results.txt. For this case, as in Case 2,
* the display field in rp is set to FALSE, preventing
* image display.
* Case 2:
* The second arg is "generate". This will cause
* generation of new golden files for the reg test.
* The results of the reg test are not recorded, and
* the display field in rp is set to FALSE.
* Case 3:
* The second arg is "display". The test will run and
* files will be written. Comparisons with golden files
* will not be carried out, so the only notion of success
* or failure is with tests that do not involve golden files.
* The display field in rp is TRUE, and this is used by
* pixDisplayWithTitle().
* (2) See regutils.h for examples of usage.
*/
l_int32
regTestSetup(l_int32 argc,
char **argv,
L_REGPARAMS **prp) {
char *testname, *vers;
char errormsg[64];
L_REGPARAMS *rp;
PROCNAME("regTestSetup");
if (argc != 1 && argc != 2) {
snprintf(errormsg, sizeof(errormsg),
"Syntax: %s [ [generate] | compare | display ]", argv[0]);
return ERROR_INT(errormsg, procName, 1);
}
if ((testname = getRootNameFromArgv0(argv[0])) == NULL)
return ERROR_INT("invalid root", procName, 1);
if ((rp = (L_REGPARAMS *) CALLOC(1, sizeof(L_REGPARAMS))) == NULL)
return ERROR_INT("rp not made", procName, 1);
*prp = rp;
rp->testname = testname;
rp->index = -1; /* increment before each test */
/* Initialize to true. A failure in any test is registered
* as a failure of the regression test. */
rp->success = TRUE;
/* Make sure the regout subdirectory exists */
lept_mkdir("regout");
/* Only open a stream to a temp file for the 'compare' case */
if (argc == 1 || !strcmp(argv[1], "compare")) {
rp->mode = L_REG_COMPARE;
rp->tempfile = genPathname("/tmp/regout", "regtest_output.txt");
rp->fp = fopenWriteStream(rp->tempfile, "wb");
if (rp->fp == NULL) {
rp->success = FALSE;
return ERROR_INT("stream not opened for tempfile", procName, 1);
}
} else if (!strcmp(argv[1], "generate")) {
rp->mode = L_REG_GENERATE;
lept_mkdir("golden");
} else if (!strcmp(argv[1], "display")) {
rp->mode = L_REG_DISPLAY;
rp->display = TRUE;
} else {
FREE(rp);
snprintf(errormsg, sizeof(errormsg),
"Syntax: %s [ [generate] | compare | display ]", argv[0]);
return ERROR_INT(errormsg, procName, 1);
}
/* Print out test name and both the leptonica and
* image libarary versions */
fprintf(stderr, "\n################ %s_reg ###############\n",
rp->testname);
vers = getLeptonicaVersion();
fprintf(stderr, "%s\n", vers);
FREE(vers);
vers = getImagelibVersions();
fprintf(stderr, "%s\n", vers);
FREE(vers);
rp->tstart = startTimerNested();
return 0;
}
int main(int argc, char **argv)
{
fprintf(stdout, "Hello World");
const char *language = "eng";
const char *datapath = "./tessdata/";
const char* filename = NULL;
PIX *pix;
tesseract::TessBaseAPI *tAPI = new tesseract::TessBaseAPI();
printf("Tesseract-ocr version: %s\n",
tesseract::TessBaseAPI::Version());
printf("Leptonica version: %s\n",
getLeptonicaVersion());
if (argv[1] == NULL) {
filename = "../phototest.tif";
} else {
filename = argv[1];
}
// Create pix
FILE *fp = fopen(filename, "r");
if (fp) {
fclose(fp);
pix = pixRead(filename);
if (pix == NULL) {
printf("Unsupported image type.\n");
exit(3);
}
else {
printf("Pix created\n");
}
} else {
printf("file '%s' doesn't exist.\n", filename);
exit(3);
}
// Initalize tesseract
int rc = tAPI->Init(datapath, language, tesseract::OEM_DEFAULT, NULL, 0, NULL, NULL, false);
if (rc) {
fprintf(stderr, "Could not initialize tesseract.\n");
exit(1);
}
// Set PIX for OCR
tAPI->SetImage(pix);
// Optional: set rectangle to OCR
//BUILD OF MATERIALS section when rendered at 600 zoom
int x1 = 5086, y1 = 175, x2 = 6835, y2 = 4165;
tAPI->SetRectangle(x1, y1, x2-x1, y2-y1);
// run ocr
char* outText = tAPI->GetUTF8Text();
printf("OCR output:\n\n");
printf(outText);
tAPI->Clear();
tAPI->End();
delete [] outText;
pixDestroy(&pix);
return 0;
}
int main(int argc, char **argv) {
#ifdef USING_GETTEXT
setlocale (LC_ALL, "");
bindtextdomain (PACKAGE, LOCALEDIR);
textdomain (PACKAGE);
#endif
if ((argc == 2 && strcmp(argv[1], "-v") == 0) ||
(argc == 2 && strcmp(argv[1], "--version") == 0)) {
char *versionStrP;
fprintf(stderr, "tesseract %s\n", tesseract::TessBaseAPI::Version());
versionStrP = getLeptonicaVersion();
fprintf(stderr, " %s\n", versionStrP);
lept_free(versionStrP);
versionStrP = getImagelibVersions();
fprintf(stderr, " %s\n", versionStrP);
lept_free(versionStrP);
exit(0);
}
tesseract::TessBaseAPI api;
int rc = api.Init(argv[0], NULL);
if (rc) {
fprintf(stderr, "Could not initialize tesseract.\n");
exit(1);
}
if (argc == 2 && strcmp(argv[1], "--list-langs") == 0) {
GenericVector<STRING> languages;
api.GetAvailableLanguagesAsVector(&languages);
fprintf(stderr, "List of available languages (%d):\n", languages.size());
for (int index = 0; index < languages.size(); ++index) {
STRING& string = languages[index];
fprintf(stderr, "%s\n", string.string());
}
api.Clear();
exit(0);
}
api.End();
// Make the order of args a bit more forgiving than it used to be.
const char* lang = "eng";
const char* image = NULL;
const char* output = NULL;
tesseract::PageSegMode pagesegmode = tesseract::PSM_AUTO;
int arg = 1;
while (arg < argc && (output == NULL || argv[arg][0] == '-')) {
if (strcmp(argv[arg], "-l") == 0 && arg + 1 < argc) {
lang = argv[arg + 1];
++arg;
} else if (strcmp(argv[arg], "-psm") == 0 && arg + 1 < argc) {
pagesegmode = static_cast<tesseract::PageSegMode>(atoi(argv[arg + 1]));
++arg;
} else if (image == NULL) {
image = argv[arg];
} else if (output == NULL) {
output = argv[arg];
}
++arg;
}
if (output == NULL) {
fprintf(stderr, _("Usage:%s imagename outputbase [-l lang] "
"[-psm pagesegmode] [configfile...]\n\n"), argv[0]);
fprintf(stderr,
_("pagesegmode values are:\n"
"0 = Orientation and script detection (OSD) only.\n"
"1 = Automatic page segmentation with OSD.\n"
"2 = Automatic page segmentation, but no OSD, or OCR\n"
"3 = Fully automatic page segmentation, but no OSD. (Default)\n"
"4 = Assume a single column of text of variable sizes.\n"
"5 = Assume a single uniform block of vertically aligned text.\n"
"6 = Assume a single uniform block of text.\n"
"7 = Treat the image as a single text line.\n"
"8 = Treat the image as a single word.\n"
"9 = Treat the image as a single word in a circle.\n"
"10 = Treat the image as a single character.\n"));
fprintf(stderr, _("-l lang and/or -psm pagesegmode must occur before any"
"configfile.\n\n"));
fprintf(stderr, _("Single options:\n"));
fprintf(stderr, _(" -v --version: version info\n"));
fprintf(stderr, _(" --list-langs: list available languages for tesseract engine\n"));
exit(1);
}
api.SetOutputName(output);
STRING tessdata_dir;
truncate_path(argv[0], &tessdata_dir);
rc = api.Init(tessdata_dir.string(), lang, tesseract::OEM_DEFAULT,
&(argv[arg]), argc - arg, NULL, NULL, false);
if (rc) {
fprintf(stderr, "Could not initialize tesseract.\n");
exit(1);
}
// We have 2 possible sources of pagesegmode: a config file and
// the command line. For backwards compatability reasons, the
// default in tesseract is tesseract::PSM_SINGLE_BLOCK, but the
// default for this program is tesseract::PSM_AUTO. We will let
// the config file take priority, so the command-line default
// can take priority over the tesseract default, so we use the
// value from the command line only if the retrieved mode
// is still tesseract::PSM_SINGLE_BLOCK, indicating no change
// in any config file. Therefore the only way to force
// tesseract::PSM_SINGLE_BLOCK is from the command line.
// It would be simpler if we could set the value before Init,
// but that doesn't work.
if (api.GetPageSegMode() == tesseract::PSM_SINGLE_BLOCK)
api.SetPageSegMode(pagesegmode);
tprintf("Tesseract Open Source OCR Engine v%s with Leptonica\n",
tesseract::TessBaseAPI::Version());
FILE* fin = fopen(image, "rb");
if (fin == NULL) {
printf("Cannot open input file: %s\n", image);
exit(2);
}
fclose(fin);
PIX *pixs;
if ((pixs = pixRead(image)) == NULL) {
printf("Unsupported image type.\n");
exit(3);
}
pixDestroy(&pixs);
STRING text_out;
if (!api.ProcessPages(image, NULL, 0, &text_out)) {
printf("Error during processing.\n");
}
bool output_hocr = false;
api.GetBoolVariable("tessedit_create_hocr", &output_hocr);
bool output_box = false;
api.GetBoolVariable("tessedit_create_boxfile", &output_box);
STRING outfile = output;
outfile += output_hocr ? ".html" : output_box ? ".box" : ".txt";
FILE* fout = fopen(outfile.string(), "wb");
if (fout == NULL) {
printf("Cannot create output file %s\n", outfile.string());
exit(1);
}
fwrite(text_out.string(), 1, text_out.length(), fout);
fclose(fout);
return 0; // Normal exit
}
int main(int argc, char **argv) {
if ((argc == 2 && strcmp(argv[1], "-v") == 0) ||
(argc == 2 && strcmp(argv[1], "--version") == 0)) {
char *versionStrP;
fprintf(stderr, "tesseract %s\n", tesseract::TessBaseAPI::Version());
versionStrP = getLeptonicaVersion();
fprintf(stderr, " %s\n", versionStrP);
lept_free(versionStrP);
versionStrP = getImagelibVersions();
fprintf(stderr, " %s\n", versionStrP);
lept_free(versionStrP);
exit(0);
}
// Make the order of args a bit more forgiving than it used to be.
const char* lang = "eng";
const char* image = NULL;
const char* output = NULL;
bool noocr = false;
bool list_langs = false;
bool print_parameters = false;
tesseract::PageSegMode pagesegmode = tesseract::PSM_AUTO;
int arg = 1;
while (arg < argc && (output == NULL || argv[arg][0] == '-')) {
if (strcmp(argv[arg], "-l") == 0 && arg + 1 < argc) {
lang = argv[arg + 1];
++arg;
} else if (strcmp(argv[arg], "-psm") == 0 && arg + 1 < argc) {
pagesegmode = static_cast<tesseract::PageSegMode>(atoi(argv[arg + 1]));
++arg;
} else if (strcmp(argv[arg], "--print-parameters") == 0) {
noocr = true;
print_parameters = true;
} else if (strcmp(argv[arg], "-c") == 0 && arg + 1 < argc) {
// handled properly after api init
++arg;
} else if (image == NULL) {
image = argv[arg];
} else if (output == NULL) {
output = argv[arg];
}
++arg;
}
if (argc == 2 && strcmp(argv[1], "--list-langs") == 0) {
list_langs = true;
noocr = true;
}
if (output == NULL && noocr == false) {
fprintf(stderr, "Usage:%s imagename outputbase|stdout [-l lang] "
"[-psm pagesegmode] [-c configvar=value] "
"[configfile...]\n\n", argv[0]);
fprintf(stderr,
"pagesegmode values are:\n"
"0 = Orientation and script detection (OSD) only.\n"
"1 = Automatic page segmentation with OSD.\n"
"2 = Automatic page segmentation, but no OSD, or OCR\n"
"3 = Fully automatic page segmentation, but no OSD. (Default)\n"
"4 = Assume a single column of text of variable sizes.\n"
"5 = Assume a single uniform block of vertically aligned text.\n"
"6 = Assume a single uniform block of text.\n"
"7 = Treat the image as a single text line.\n"
"8 = Treat the image as a single word.\n"
"9 = Treat the image as a single word in a circle.\n"
"10 = Treat the image as a single character.\n");
fprintf(stderr, "multiple -c arguments are allowed.\n");
fprintf(stderr, "-l lang, -psm pagesegmode and any -c options must occur"
"before any configfile.\n\n");
fprintf(stderr, "Single options:\n");
fprintf(stderr, " -v --version: version info\n");
fprintf(stderr, " --list-langs: list available languages for tesseract "
"engine\n");
fprintf(stderr, " --print-parameters: print tesseract parameters to the "
"stdout\n");
exit(1);
}
tesseract::TessBaseAPI api;
STRING tessdata_dir;
truncate_path(argv[0], &tessdata_dir);
api.SetOutputName(output);
int rc = api.Init(tessdata_dir.string(), lang, tesseract::OEM_DEFAULT,
&(argv[arg]), argc - arg, NULL, NULL, false);
if (rc) {
fprintf(stderr, "Could not initialize tesseract.\n");
exit(1);
}
char opt1[255], opt2[255];
for (arg = 0; arg < argc; arg++) {
if (strcmp(argv[arg], "-c") == 0 && arg + 1 < argc) {
strncpy(opt1, argv[arg + 1], 255);
*(strchr(opt1, '=')) = 0;
strncpy(opt2, strchr(argv[arg + 1], '=') + 1, 255);
opt2[254] = 0;
++arg;
if(!api.SetVariable(opt1, opt2)) {
fprintf(stderr, "Could not set option: %s=%s\n", opt1, opt2);
}
}
}
if (list_langs) {
GenericVector<STRING> languages;
api.GetAvailableLanguagesAsVector(&languages);
fprintf(stderr, "List of available languages (%d):\n",
languages.size());
for (int index = 0; index < languages.size(); ++index) {
STRING& string = languages[index];
fprintf(stderr, "%s\n", string.string());
}
api.End();
exit(0);
}
if (print_parameters) {
FILE* fout = stdout;
fprintf(stdout, "Tesseract parameters:\n");
api.PrintVariables(fout);
api.End();
exit(0);
}
// We have 2 possible sources of pagesegmode: a config file and
// the command line. For backwards compatability reasons, the
// default in tesseract is tesseract::PSM_SINGLE_BLOCK, but the
// default for this program is tesseract::PSM_AUTO. We will let
// the config file take priority, so the command-line default
// can take priority over the tesseract default, so we use the
// value from the command line only if the retrieved mode
// is still tesseract::PSM_SINGLE_BLOCK, indicating no change
// in any config file. Therefore the only way to force
// tesseract::PSM_SINGLE_BLOCK is from the command line.
// It would be simpler if we could set the value before Init,
// but that doesn't work.
if (api.GetPageSegMode() == tesseract::PSM_SINGLE_BLOCK)
api.SetPageSegMode(pagesegmode);
tprintf("Tesseract Open Source OCR Engine v%s with Leptonica\n",
tesseract::TessBaseAPI::Version());
FILE* fin = fopen(image, "rb");
if (fin == NULL) {
fprintf(stderr, "Cannot open input file: %s\n", image);
exit(2);
}
fclose(fin);
bool output_hocr = false;
api.GetBoolVariable("tessedit_create_hocr", &output_hocr);
bool output_box = false;
api.GetBoolVariable("tessedit_create_boxfile", &output_box);
FILE* fout = stdout;
if (strcmp(output, "-") && strcmp(output, "stdout")) {
STRING outfile = output;
outfile += output_hocr ? ".html" : output_box ? ".box" : ".txt";
fout = fopen(outfile.string(), "wb");
if (fout == NULL) {
fprintf(stderr, "Cannot create output file %s\n", outfile.string());
exit(1);
}
}
STRING text_out;
if (!api.ProcessPages(image, NULL, 0, &text_out)) {
fprintf(stderr, "Error during processing.\n");
if (fout != stdout)
fclose(fout);
exit(1);
}
fwrite(text_out.string(), 1, text_out.length(), fout);
if (fout != stdout)
fclose(fout);
else
clearerr(fout);
return 0; // Normal exit
}
main(int argc,
char **argv)
{
char *str, *results_file;
char command[256], buf[256];
l_int32 i, ntests, dotest, nfail, ret, start, stop;
SARRAY *sa;
L_TIMER timer;
static char mainName[] = "alltests_reg";
if (argc != 2)
return ERROR_INT(" Syntax alltests_reg [generate | compare | display]",
mainName, 1);
l_getCurrentTime(&start, NULL);
ntests = sizeof(tests) / sizeof(char *);
fprintf(stderr, "Running alltests_reg:\n"
"This currently tests %d of the 97 Regression Test\n"
"programs in the /prog directory.\n", ntests);
/* Clear the output file if we're doing the set of reg tests */
dotest = strcmp(argv[1], "compare") ? 0 : 1;
if (dotest) {
results_file = genPathname("/tmp", "reg_results.txt");
sa = sarrayCreate(3);
sarrayAddString(sa, (char *)header, L_COPY);
sarrayAddString(sa, getLeptonicaVersion(), L_INSERT);
sarrayAddString(sa, getImagelibVersions(), L_INSERT);
str = sarrayToString(sa, 1);
sarrayDestroy(&sa);
l_binaryWrite("/tmp/reg_results.txt", "w", str, strlen(str));
lept_free(str);
}
nfail = 0;
for (i = 0; i < ntests; i++) {
snprintf(command, sizeof(command) - 2, "./%s %s", tests[i], argv[1]);
ret = system(command);
if (ret) {
snprintf(buf, sizeof(buf), "Failed to complete %s\n", tests[i]);
if (dotest) {
l_binaryWrite("/tmp/reg_results.txt", "a", buf, strlen(buf));
nfail++;
}
else
fprintf(stderr, "%s", buf);
}
}
if (dotest) {
#ifndef _WIN32
snprintf(command, sizeof(command) - 2, "cat %s", results_file);
#else
snprintf(command, sizeof(command) - 2, "type \"%s\"", results_file);
#endif /* !_WIN32 */
lept_free(results_file);
ret = system(command);
fprintf(stderr, "Success in %d of %d *_reg programs (output matches"
" the \"golden\" files)\n", ntests - nfail, ntests);
}
l_getCurrentTime(&stop, NULL);
fprintf(stderr, "Time for all regression tests: %d sec\n", stop - start);
return 0;
}
/*!
* pixWriteStreamJp2k()
*
* Input: stream
* pix (any depth, cmap is OK)
* quality (SNR > 0; default ~34; 0 for lossless encoding)
* nlevels (<= 10)
* hint (a bitwise OR of L_JP2K_* values; 0 for default)
* debug (output callback messages, etc)
* Return: 0 if OK, 1 on error
* Notes:
* (1) See pixWriteJp2k() for usage.
* (2) For an encoder with more encoding options, see, e.g.,
* https://github.com/OpenJPEG/openjpeg/blob/master/tests/test_tile_encoder.c
*/
l_int32
pixWriteStreamJp2k(FILE *fp,
PIX *pix,
l_int32 quality,
l_int32 nlevels,
l_int32 hint,
l_int32 debug)
{
l_int32 w, h, d, success, snr;
const char *opjVersion;
PIX *pixs;
opj_cparameters_t parameters; /* compression parameters */
opj_stream_t *l_stream = NULL;
opj_codec_t* l_codec = NULL;;
opj_image_t *image = NULL;
PROCNAME("pixWriteStreamJp2k");
if (!fp)
return ERROR_INT("stream not open", procName, 1);
if (!pix)
return ERROR_INT("pix not defined", procName, 1);
if (quality < 0)
return ERROR_INT("quality must be >= 0", procName, 1);
if (quality > 0 && quality < 27)
L_WARNING("SNR = %d < 27; very low\n", procName, quality);
if (quality > 45)
L_WARNING("SNR = %d > 45; nearly lossless\n", procName, quality);
snr = (l_float32)quality;
if (nlevels <= 0) nlevels = 5; /* default */
if (nlevels > 10) {
L_WARNING("nlevels = %d > 10; setting to 10\n", procName, nlevels);
nlevels = 10;
}
opjVersion = opj_version();
if (opjVersion[0] != '2') {
L_ERROR("version is %s; must be 2.0 or higher\n", procName, opjVersion);
return 1;
}
if ((opjVersion[2] - 0x30) != OPJ_VERSION_MINOR) {
L_ERROR("version %s: differs from minor = %d\n",
procName, opjVersion, OPJ_VERSION_MINOR);
return 1;
}
/* Remove colormap if it exists; result is 8 or 32 bpp */
pixGetDimensions(pix, &w, &h, &d);
if (d == 24) {
pixs = pixConvert24To32(pix);
} else if (d == 32) {
pixs = pixClone(pix);
} else if (pixGetColormap(pix) == NULL) {
pixs = pixConvertTo8(pix, 0);
} else { /* colormap */
L_INFO("removing colormap; may be better to compress losslessly\n",
procName);
pixs = pixRemoveColormap(pix, REMOVE_CMAP_BASED_ON_SRC);
}
/* Convert to opj image format. */
image = pixConvertToOpjImage(pixs);
pixDestroy(&pixs);
/* Set encoding parameters to default values.
* We use one layer with the input SNR. */
opj_set_default_encoder_parameters(¶meters);
parameters.cp_fixed_quality = 1;
parameters.cp_disto_alloc = 0;
parameters.cp_fixed_alloc = 0;
parameters.tcp_distoratio[0] = snr;
parameters.tcp_numlayers = 1;
parameters.numresolution = nlevels + 1;
/* Create comment for codestream */
if (parameters.cp_comment == NULL) {
const char comment1[] = "Created by Leptonica, version ";
const char comment2[] = "; using OpenJPEG, version ";
size_t len1 = strlen(comment1);
size_t len2 = strlen(comment2);
char *version1 = getLeptonicaVersion();
const char *version2 = opj_version();
len1 += len2 + strlen(version1) + strlen(version2) + 1;
parameters.cp_comment = (char *)MALLOC(len1);
snprintf(parameters.cp_comment, len1, "%s%s%s%s", comment1, version1,
comment2, version2);
FREE(version1);
}
/* Get the encoder handle */
if ((l_codec = opj_create_compress(OPJ_CODEC_JP2)) == NULL) {
opj_image_destroy(image);
FREE(parameters.cp_comment);
return ERROR_INT("failed to get the encoder handle\n", procName, 1);
}
/* Catch and report events using callbacks */
if (debug) {
opj_set_info_handler(l_codec, info_callback, NULL);
opj_set_warning_handler(l_codec, warning_callback, NULL);
opj_set_error_handler(l_codec, error_callback, NULL);
}
/* Set up the encoder */
if (!opj_setup_encoder(l_codec, ¶meters, image)) {
opj_destroy_codec(l_codec);
opj_image_destroy(image);
FREE(parameters.cp_comment);
return ERROR_INT("failed to set up the encoder\n", procName, 1);
}
/* Open a compression stream for writing. In 2.0 we could use this:
* opj_stream_create_default_file_stream(fp, 0)
* but the file stream interface was removed in 2.1. */
rewind(fp);
if ((l_stream = opjCreateStream(fp, 0)) == NULL) {
opj_destroy_codec(l_codec);
opj_image_destroy(image);
FREE(parameters.cp_comment);
return ERROR_INT("failed to open l_stream\n", procName, 1);
}
/* Encode the image */
if (!opj_start_compress(l_codec, image, l_stream)) {
opj_stream_destroy(l_stream);
opj_destroy_codec(l_codec);
opj_image_destroy(image);
FREE(parameters.cp_comment);
return ERROR_INT("opj_start_compress failed\n", procName, 1);
}
if (!opj_encode(l_codec, l_stream)) {
opj_stream_destroy(l_stream);
opj_destroy_codec(l_codec);
opj_image_destroy(image);
FREE(parameters.cp_comment);
return ERROR_INT("opj_encode failed\n", procName, 1);
}
success = opj_end_compress(l_codec, l_stream);
/* Clean up */
opj_stream_destroy(l_stream);
opj_destroy_codec(l_codec);
opj_image_destroy(image);
FREE(parameters.cp_comment);
if (success)
return 0;
else
return ERROR_INT("opj_end_compress failed\n", procName, 1);
}