tesseract  4.0.0-1-g2a2b
tesseract::ImageThresholder Class Reference

#include <thresholder.h>

Public Member Functions

 ImageThresholder ()
 
virtual ~ImageThresholder ()
 
virtual void Clear ()
 Destroy the Pix if there is one, freeing memory. More...
 
bool IsEmpty () const
 Return true if no image has been set. More...
 
void SetImage (const unsigned char *imagedata, int width, int height, int bytes_per_pixel, int bytes_per_line)
 
void SetRectangle (int left, int top, int width, int height)
 
virtual void GetImageSizes (int *left, int *top, int *width, int *height, int *imagewidth, int *imageheight)
 
bool IsColor () const
 Return true if the source image is color. More...
 
bool IsBinary () const
 Returns true if the source image is binary. More...
 
int GetScaleFactor () const
 
void SetSourceYResolution (int ppi)
 
int GetSourceYResolution () const
 
int GetScaledYResolution () const
 
void SetEstimatedResolution (int ppi)
 
int GetScaledEstimatedResolution () const
 
void SetImage (const Pix *pix)
 
virtual bool ThresholdToPix (PageSegMode pageseg_mode, Pix **pix)
 Returns false on error. More...
 
virtual Pix * GetPixRectThresholds ()
 
Pix * GetPixRect ()
 
virtual Pix * GetPixRectGrey ()
 

Protected Member Functions

virtual void Init ()
 Common initialization shared between SetImage methods. More...
 
bool IsFullImage () const
 Return true if we are processing the full image. More...
 
void OtsuThresholdRectToPix (Pix *src_pix, Pix **out_pix) const
 
void ThresholdRectToPix (Pix *src_pix, int num_channels, const int *thresholds, const int *hi_values, Pix **pix) const
 

Protected Attributes

Pix * pix_
 
int image_width_
 
int image_height_
 
int pix_channels_
 
int pix_wpl_
 
int scale_
 
int yres_
 
int estimated_res_
 
int rect_left_
 
int rect_top_
 
int rect_width_
 
int rect_height_
 

Detailed Description

Base class for all tesseract image thresholding classes. Specific classes can add new thresholding methods by overriding ThresholdToPix. Each instance deals with a single image, but the design is intended to be useful for multiple calls to SetRectangle and ThresholdTo* if desired.

Definition at line 36 of file thresholder.h.

Constructor & Destructor Documentation

◆ ImageThresholder()

tesseract::ImageThresholder::ImageThresholder ( )

Definition at line 33 of file thresholder.cpp.

◆ ~ImageThresholder()

tesseract::ImageThresholder::~ImageThresholder ( )
virtual

Definition at line 41 of file thresholder.cpp.

41  {
42  Clear();
43 }
virtual void Clear()
Destroy the Pix if there is one, freeing memory.
Definition: thresholder.cpp:46

Member Function Documentation

◆ Clear()

void tesseract::ImageThresholder::Clear ( )
virtual

Destroy the Pix if there is one, freeing memory.

Definition at line 46 of file thresholder.cpp.

46  {
47  pixDestroy(&pix_);
48 }

◆ GetImageSizes()

void tesseract::ImageThresholder::GetImageSizes ( int *  left,
int *  top,
int *  width,
int *  height,
int *  imagewidth,
int *  imageheight 
)
virtual

Get enough parameters to be able to rebuild bounding boxes in the original image (not just within the rectangle). Left and top are enough with top-down coordinates, but the height of the rectangle and the image are needed for bottom-up.

Definition at line 133 of file thresholder.cpp.

135  {
136  *left = rect_left_;
137  *top = rect_top_;
138  *width = rect_width_;
139  *height = rect_height_;
140  *imagewidth = image_width_;
141  *imageheight = image_height_;
142 }

◆ GetPixRect()

Pix * tesseract::ImageThresholder::GetPixRect ( )

Get a clone/copy of the source image rectangle. The returned Pix must be pixDestroyed. This function will be used in the future by the page layout analysis, and the layout analysis that uses it will only be available with Leptonica, so there is no raw equivalent.

Definition at line 235 of file thresholder.cpp.

235  {
236  if (IsFullImage()) {
237  // Just clone the whole thing.
238  return pixClone(pix_);
239  } else {
240  // Crop to the given rectangle.
241  Box* box = boxCreate(rect_left_, rect_top_, rect_width_, rect_height_);
242  Pix* cropped = pixClipRectangle(pix_, box, nullptr);
243  boxDestroy(&box);
244  return cropped;
245  }
246 }
bool IsFullImage() const
Return true if we are processing the full image.
Definition: thresholder.h:153

◆ GetPixRectGrey()

Pix * tesseract::ImageThresholder::GetPixRectGrey ( )
virtual

Definition at line 252 of file thresholder.cpp.

252  {
253  Pix* pix = GetPixRect(); // May have to be reduced to grey.
254  int depth = pixGetDepth(pix);
255  if (depth != 8) {
256  Pix* result = depth < 8 ? pixConvertTo8(pix, false)
257  : pixConvertRGBToLuminance(pix);
258  pixDestroy(&pix);
259  return result;
260  }
261  return pix;
262 }

◆ GetPixRectThresholds()

Pix * tesseract::ImageThresholder::GetPixRectThresholds ( )
virtual

Definition at line 208 of file thresholder.cpp.

208  {
209  if (IsBinary()) return nullptr;
210  Pix* pix_grey = GetPixRectGrey();
211  int width = pixGetWidth(pix_grey);
212  int height = pixGetHeight(pix_grey);
213  int* thresholds;
214  int* hi_values;
215  OtsuThreshold(pix_grey, 0, 0, width, height, &thresholds, &hi_values);
216  pixDestroy(&pix_grey);
217  Pix* pix_thresholds = pixCreate(width, height, 8);
218  int threshold = thresholds[0] > 0 ? thresholds[0] : 128;
219  pixSetAllArbitrary(pix_thresholds, threshold);
220  delete [] thresholds;
221  delete [] hi_values;
222  return pix_thresholds;
223 }
virtual Pix * GetPixRectGrey()
bool IsBinary() const
Returns true if the source image is binary.
Definition: thresholder.h:75
int OtsuThreshold(Pix *src_pix, int left, int top, int width, int height, int **thresholds, int **hi_values)
Definition: otsuthr.cpp:39

◆ GetScaledEstimatedResolution()

int tesseract::ImageThresholder::GetScaledEstimatedResolution ( ) const
inline

Definition at line 106 of file thresholder.h.

106  {
107  return scale_ * estimated_res_;
108  }

◆ GetScaledYResolution()

int tesseract::ImageThresholder::GetScaledYResolution ( ) const
inline

Definition at line 93 of file thresholder.h.

93  {
94  return scale_ * yres_;
95  }

◆ GetScaleFactor()

int tesseract::ImageThresholder::GetScaleFactor ( ) const
inline

Definition at line 79 of file thresholder.h.

79  {
80  return scale_;
81  }

◆ GetSourceYResolution()

int tesseract::ImageThresholder::GetSourceYResolution ( ) const
inline

Definition at line 90 of file thresholder.h.

90  {
91  return yres_;
92  }

◆ Init()

void tesseract::ImageThresholder::Init ( )
protectedvirtual

Common initialization shared between SetImage methods.

Definition at line 226 of file thresholder.cpp.

226  {
228 }
void SetRectangle(int left, int top, int width, int height)

◆ IsBinary()

bool tesseract::ImageThresholder::IsBinary ( ) const
inline

Returns true if the source image is binary.

Definition at line 75 of file thresholder.h.

75  {
76  return pix_channels_ == 0;
77  }

◆ IsColor()

bool tesseract::ImageThresholder::IsColor ( ) const
inline

Return true if the source image is color.

Definition at line 70 of file thresholder.h.

70  {
71  return pix_channels_ >= 3;
72  }

◆ IsEmpty()

bool tesseract::ImageThresholder::IsEmpty ( ) const

Return true if no image has been set.

Definition at line 51 of file thresholder.cpp.

51  {
52  return pix_ == nullptr;
53 }

◆ IsFullImage()

bool tesseract::ImageThresholder::IsFullImage ( ) const
inlineprotected

Return true if we are processing the full image.

Definition at line 153 of file thresholder.h.

◆ OtsuThresholdRectToPix()

void tesseract::ImageThresholder::OtsuThresholdRectToPix ( Pix *  src_pix,
Pix **  out_pix 
) const
protected

Definition at line 265 of file thresholder.cpp.

266  {
267  PERF_COUNT_START("OtsuThresholdRectToPix")
268  int* thresholds;
269  int* hi_values;
270 
271  int num_channels = OtsuThreshold(src_pix, rect_left_, rect_top_, rect_width_,
272  rect_height_, &thresholds, &hi_values);
273  // only use opencl if compiled w/ OpenCL and selected device is opencl
274 #ifdef USE_OPENCL
275  OpenclDevice od;
276  if (num_channels == 4 &&
277  od.selectedDeviceIsOpenCL() && rect_top_ == 0 && rect_left_ == 0) {
278  od.ThresholdRectToPixOCL((unsigned char*)pixGetData(src_pix), num_channels,
279  pixGetWpl(src_pix) * 4, thresholds, hi_values,
280  out_pix /*pix_OCL*/, rect_height_, rect_width_,
282  } else {
283 #endif
284  ThresholdRectToPix(src_pix, num_channels, thresholds, hi_values, out_pix);
285 #ifdef USE_OPENCL
286  }
287 #endif
288  delete [] thresholds;
289  delete [] hi_values;
290 
292 }
#define PERF_COUNT_START(FUNCT_NAME)
void ThresholdRectToPix(Pix *src_pix, int num_channels, const int *thresholds, const int *hi_values, Pix **pix) const
#define PERF_COUNT_END
int OtsuThreshold(Pix *src_pix, int left, int top, int width, int height, int **thresholds, int **hi_values)
Definition: otsuthr.cpp:39

◆ SetEstimatedResolution()

void tesseract::ImageThresholder::SetEstimatedResolution ( int  ppi)
inline

Definition at line 101 of file thresholder.h.

101  {
102  estimated_res_ = ppi;
103  }

◆ SetImage() [1/2]

void tesseract::ImageThresholder::SetImage ( const unsigned char *  imagedata,
int  width,
int  height,
int  bytes_per_pixel,
int  bytes_per_line 
)

SetImage makes a copy of all the image data, so it may be deleted immediately after this call. Greyscale of 8 and color of 24 or 32 bits per pixel may be given. Palette color images will not work properly and must be converted to 24 bit. Binary images of 1 bit per pixel may also be given but they must be byte packed with the MSB of the first byte being the first pixel, and a one pixel is WHITE. For binary images set bytes_per_pixel=0.

Definition at line 63 of file thresholder.cpp.

65  {
66  int bpp = bytes_per_pixel * 8;
67  if (bpp == 0) bpp = 1;
68  Pix* pix = pixCreate(width, height, bpp == 24 ? 32 : bpp);
69  l_uint32* data = pixGetData(pix);
70  int wpl = pixGetWpl(pix);
71  switch (bpp) {
72  case 1:
73  for (int y = 0; y < height; ++y, data += wpl, imagedata += bytes_per_line) {
74  for (int x = 0; x < width; ++x) {
75  if (imagedata[x / 8] & (0x80 >> (x % 8)))
76  CLEAR_DATA_BIT(data, x);
77  else
78  SET_DATA_BIT(data, x);
79  }
80  }
81  break;
82 
83  case 8:
84  // Greyscale just copies the bytes in the right order.
85  for (int y = 0; y < height; ++y, data += wpl, imagedata += bytes_per_line) {
86  for (int x = 0; x < width; ++x)
87  SET_DATA_BYTE(data, x, imagedata[x]);
88  }
89  break;
90 
91  case 24:
92  // Put the colors in the correct places in the line buffer.
93  for (int y = 0; y < height; ++y, imagedata += bytes_per_line) {
94  for (int x = 0; x < width; ++x, ++data) {
95  SET_DATA_BYTE(data, COLOR_RED, imagedata[3 * x]);
96  SET_DATA_BYTE(data, COLOR_GREEN, imagedata[3 * x + 1]);
97  SET_DATA_BYTE(data, COLOR_BLUE, imagedata[3 * x + 2]);
98  }
99  }
100  break;
101 
102  case 32:
103  // Maintain byte order consistency across different endianness.
104  for (int y = 0; y < height; ++y, imagedata += bytes_per_line, data += wpl) {
105  for (int x = 0; x < width; ++x) {
106  data[x] = (imagedata[x * 4] << 24) | (imagedata[x * 4 + 1] << 16) |
107  (imagedata[x * 4 + 2] << 8) | imagedata[x * 4 + 3];
108  }
109  }
110  break;
111 
112  default:
113  tprintf("Cannot convert RAW image to Pix with bpp = %d\n", bpp);
114  }
115  pixSetYRes(pix, 300);
116  SetImage(pix);
117  pixDestroy(&pix);
118 }
void SetImage(const unsigned char *imagedata, int width, int height, int bytes_per_pixel, int bytes_per_line)
Definition: thresholder.cpp:63
DLLSYM void tprintf(const char *format,...)
Definition: tprintf.cpp:37

◆ SetImage() [2/2]

void tesseract::ImageThresholder::SetImage ( const Pix *  pix)

Pix vs raw, which to use? Pix is the preferred input for efficiency, since raw buffers are copied. SetImage for Pix clones its input, so the source pix may be pixDestroyed immediately after, but may not go away until after the Thresholder has finished with it.

Definition at line 149 of file thresholder.cpp.

149  {
150  if (pix_ != nullptr)
151  pixDestroy(&pix_);
152  Pix* src = const_cast<Pix*>(pix);
153  int depth;
154  pixGetDimensions(src, &image_width_, &image_height_, &depth);
155  // Convert the image as necessary so it is one of binary, plain RGB, or
156  // 8 bit with no colormap. Guarantee that we always end up with our own copy,
157  // not just a clone of the input.
158  if (pixGetColormap(src)) {
159  Pix* tmp = pixRemoveColormap(src, REMOVE_CMAP_BASED_ON_SRC);
160  depth = pixGetDepth(tmp);
161  if (depth > 1 && depth < 8) {
162  pix_ = pixConvertTo8(tmp, false);
163  pixDestroy(&tmp);
164  } else {
165  pix_ = tmp;
166  }
167  } else if (depth > 1 && depth < 8) {
168  pix_ = pixConvertTo8(src, false);
169  } else {
170  pix_ = pixCopy(nullptr, src);
171  }
172  depth = pixGetDepth(pix_);
173  pix_channels_ = depth / 8;
174  pix_wpl_ = pixGetWpl(pix_);
175  scale_ = 1;
176  estimated_res_ = yres_ = pixGetYRes(pix_);
177  Init();
178 }
virtual void Init()
Common initialization shared between SetImage methods.

◆ SetRectangle()

void tesseract::ImageThresholder::SetRectangle ( int  left,
int  top,
int  width,
int  height 
)

Store the coordinates of the rectangle to process for later use. Doesn't actually do any thresholding.

Definition at line 122 of file thresholder.cpp.

122  {
123  rect_left_ = left;
124  rect_top_ = top;
125  rect_width_ = width;
126  rect_height_ = height;
127 }

◆ SetSourceYResolution()

void tesseract::ImageThresholder::SetSourceYResolution ( int  ppi)
inline

Definition at line 86 of file thresholder.h.

86  {
87  yres_ = ppi;
88  estimated_res_ = ppi;
89  }

◆ ThresholdRectToPix()

void tesseract::ImageThresholder::ThresholdRectToPix ( Pix *  src_pix,
int  num_channels,
const int *  thresholds,
const int *  hi_values,
Pix **  pix 
) const
protected

Threshold the rectangle, taking everything except the src_pix from the class, using thresholds/hi_values to the output pix. NOTE that num_channels is the size of the thresholds and hi_values

Definition at line 298 of file thresholder.cpp.

302  {
303  PERF_COUNT_START("ThresholdRectToPix")
304  *pix = pixCreate(rect_width_, rect_height_, 1);
305  uint32_t* pixdata = pixGetData(*pix);
306  int wpl = pixGetWpl(*pix);
307  int src_wpl = pixGetWpl(src_pix);
308  uint32_t* srcdata = pixGetData(src_pix);
309  for (int y = 0; y < rect_height_; ++y) {
310  const uint32_t* linedata = srcdata + (y + rect_top_) * src_wpl;
311  uint32_t* pixline = pixdata + y * wpl;
312  for (int x = 0; x < rect_width_; ++x) {
313  bool white_result = true;
314  for (int ch = 0; ch < num_channels; ++ch) {
315  int pixel =
316  GET_DATA_BYTE(linedata, (x + rect_left_) * num_channels + ch);
317  if (hi_values[ch] >= 0 &&
318  (pixel > thresholds[ch]) == (hi_values[ch] == 0)) {
319  white_result = false;
320  break;
321  }
322  }
323  if (white_result)
324  CLEAR_DATA_BIT(pixline, x);
325  else
326  SET_DATA_BIT(pixline, x);
327  }
328  }
329 
331 }
#define PERF_COUNT_START(FUNCT_NAME)
#define PERF_COUNT_END

◆ ThresholdToPix()

bool tesseract::ImageThresholder::ThresholdToPix ( PageSegMode  pageseg_mode,
Pix **  pix 
)
virtual

Returns false on error.

Threshold the source image as efficiently as possible to the output Pix. Creates a Pix and sets pix to point to the resulting pointer. Caller must use pixDestroy to free the created Pix. Returns false on error.

Definition at line 184 of file thresholder.cpp.

184  {
185  if (image_width_ > INT16_MAX || image_height_ > INT16_MAX) {
186  tprintf("Image too large: (%d, %d)\n", image_width_, image_height_);
187  return false;
188  }
189  if (pix_channels_ == 0) {
190  // We have a binary image, but it still has to be copied, as this API
191  // allows the caller to modify the output.
192  Pix* original = GetPixRect();
193  *pix = pixCopy(nullptr, original);
194  pixDestroy(&original);
195  } else {
197  }
198  return true;
199 }
void OtsuThresholdRectToPix(Pix *src_pix, Pix **out_pix) const
DLLSYM void tprintf(const char *format,...)
Definition: tprintf.cpp:37

Member Data Documentation

◆ estimated_res_

int tesseract::ImageThresholder::estimated_res_
protected

Definition at line 181 of file thresholder.h.

◆ image_height_

int tesseract::ImageThresholder::image_height_
protected

Definition at line 175 of file thresholder.h.

◆ image_width_

int tesseract::ImageThresholder::image_width_
protected

Definition at line 174 of file thresholder.h.

◆ pix_

Pix* tesseract::ImageThresholder::pix_
protected

Clone or other copy of the source Pix. The pix will always be PixDestroy()ed on destruction of the class.

Definition at line 172 of file thresholder.h.

◆ pix_channels_

int tesseract::ImageThresholder::pix_channels_
protected

Definition at line 176 of file thresholder.h.

◆ pix_wpl_

int tesseract::ImageThresholder::pix_wpl_
protected

Definition at line 177 of file thresholder.h.

◆ rect_height_

int tesseract::ImageThresholder::rect_height_
protected

Definition at line 185 of file thresholder.h.

◆ rect_left_

int tesseract::ImageThresholder::rect_left_
protected

Definition at line 182 of file thresholder.h.

◆ rect_top_

int tesseract::ImageThresholder::rect_top_
protected

Definition at line 183 of file thresholder.h.

◆ rect_width_

int tesseract::ImageThresholder::rect_width_
protected

Definition at line 184 of file thresholder.h.

◆ scale_

int tesseract::ImageThresholder::scale_
protected

Definition at line 179 of file thresholder.h.

◆ yres_

int tesseract::ImageThresholder::yres_
protected

Definition at line 180 of file thresholder.h.


The documentation for this class was generated from the following files: