DcmSegUtils Class Reference

Class that contains helper functionality for the dcmseg module. More...

Static Public Member Functions
static DcmIODTypes::Frame *	packBinaryFrame (const Uint8 *pixelData, const Uint16 rows, const Uint16 columns)
	Pack the given segmentation pixel data, provided "unpacked", into the packed format expected by DICOM. More...
static size_t	getBytesForBinaryFrame (const size_t &numPixels)
	Compute the number of bytes required for a binary pixel data frame, given the number of pixels. More...
static DcmIODTypes::Frame *	unpackBinaryFrame (const DcmIODTypes::Frame *frame, Uint16 rows, Uint16 cols)
	Unpacks a binary segmentation frame into a "sparse" pixel data frame where every resulting byte represents a single bit of the frame being either 0 (not set) or 1 (set). More...
static void	alignFrameOnBitPosition (Uint8 *buf, const size_t bufLen, const Uint8 numBits)
	Aligns 1 bit per pixel frame data to make the frame start at a specific bit position within the first byte. More...
static void	alignFrameOnByteBoundary (Uint8 *buf, const size_t bufLen, const Uint8 numBits)
	Aligns 1 bit per pixel frame data starting at a given bit position in the provided buffer with the start of that buffer. More...
static OFString	debugByte2Bin (Uint8 b)
	Dumps a byte as binary number to a string. More...
static void	debugDumpBin (Uint8 *buffer, size_t length, const char *what)
	Dumps a memory block byte for byte to the debug log stream. More...

Detailed Description

Class that contains helper functionality for the dcmseg module.

Member Function Documentation

alignFrameOnBitPosition()

static void DcmSegUtils::alignFrameOnBitPosition ( Uint8 *  buf, const size_t  bufLen, const Uint8  numBits  )

static

Aligns 1 bit per pixel frame data to make the frame start at a specific bit position within the first byte.

This is used in the context that dcmseg holds the frames in memory aligned to exact byte positions, while the DICOM encoding might require a frame to start at an arbitrary bit position since all (1 bit per pixel) frames are directly concatenated one after another (i.e. if one frame does not occupy a number of bits dividable by 8, not all frames will be aligned at exact byte positions). Note that each byte is filled from the right, i.e. the first pixel will represented by the bit at the very right of the first byte, and the 9th pixel will be in the very right position of the following byte. This is not a regular bit shift operation since the bits from the previous frame are on the left of the byte, but must be aligned at the right. The current frame starts from the first bit, occupying the unused bits of the last frame and then continuing in the next byte at the first bit from the left. Example for two bit shift: Input buffer bytes: hgfedcba 87654321 Result: fedcba00 654321hg The 00 in the first byte must be handled by the caller (will contain the two bits of the previous frame). See also dcmseg/tests/tutils.cc for more examples.

Parameters

buf	The address of the memory buffer to shift
bufLen	The length of the buf memory block in bytes
numBits	The number of bits to shift. Must be 0 <= numBits <= 7.

alignFrameOnByteBoundary()

static void DcmSegUtils::alignFrameOnByteBoundary ( Uint8 *  buf, const size_t  bufLen, const Uint8  numBits  )

static

Aligns 1 bit per pixel frame data starting at a given bit position in the provided buffer with the start of that buffer.

This is used to create a frame structure where all the bytes (including the first one) only contain data from the frame at hand. Note that each byte is filled from the right, i.e. the first pixel will represented by the bit at the very right of the first byte, and the 9th pixel will be in the very right position of the following byte. Example: 3 bytes input buffer: edcbaZYX mlkjihgf utsrqpon Result after aligning 3 bits: fghedcba ponmlkji 000utsrq The 000 are unused bits and therefore zeroed out in the last byte. Bits ZYX will be shifted out which is ok since it does not belong to the current frame. See also dcmseg/tests/tutils.cc for more examples.

Parameters

buf	The address of the memory buffer to shift
bufLen	The length of the buf memory block in bytes
numBits	The number of bits to shift. Must be 0 <= numBits <= 7.

debugByte2Bin()

static OFString DcmSegUtils::debugByte2Bin ( Uint8  b )

static

Dumps a byte as binary number to a string.

Only useful for debugging purposes.

Parameters

b	The byte to dump

Returns

A string containing b as a binary number

debugDumpBin()

static void DcmSegUtils::debugDumpBin ( Uint8 *  buffer, size_t  length, const char *  what  )

static

Dumps a memory block byte for byte to the debug log stream.

Only useful for debugging purposes.

Parameters

buffer	The address of the memory block to dump
length	The length of memory to be dumped
what	String describing what is dumped.

getBytesForBinaryFrame()

static size_t DcmSegUtils::getBytesForBinaryFrame ( const size_t &  numPixels )

static

Compute the number of bytes required for a binary pixel data frame, given the number of pixels.

Parameters

numPixels

The total number of pixels

Returns

The number of bytes required to pack the data into a binary segmentation frame

packBinaryFrame()

static DcmIODTypes::Frame* DcmSegUtils::packBinaryFrame ( const Uint8 *  pixelData, const Uint16  rows, const Uint16  columns  )

static

Pack the given segmentation pixel data, provided "unpacked", into the packed format expected by DICOM.

Parameters

pixelData	Pixel data in unpacked format
rows	Number of rows in the pixel data
columns	The number of columns in the pixel data

Returns

The frame data if successful, NULL if an error occurs

unpackBinaryFrame()

static DcmIODTypes::Frame* DcmSegUtils::unpackBinaryFrame ( const DcmIODTypes::Frame *  frame, Uint16  rows, Uint16  cols  )

static

Unpacks a binary segmentation frame into a "sparse" pixel data frame where every resulting byte represents a single bit of the frame being either 0 (not set) or 1 (set).

Parameters

frame	The input buffer with the frame in packed format
rows	The rows of the frame
cols	The cols of the frame

Returns

The segmentation frame in unpacked format. NULL in case of error.

---

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

• dcmseg/include/dcmtk/dcmseg/segutils.h