ILImage Interface Reference

This interface expose APIs to access/manipulate/draw a LPV LImage object. More...

Inheritance diagram for ILImage:

Public Member Functions
void	CopyFrom (ILImage *img, ILRegion *region, int panX, int panY)
void	CopyTo (ILImage *dstImg)
void	Draw (HDC hdc, double zoomX, double zoomY, double panX, double panY)
void	DrawAsMask (HDC hdc, double zoomX, double zoomY, double panX, double panY)
ILImage *	GetChannel (int channelIdx)
LPVErrorCode	LoadRawFile (LString filename, int w, int h, int flags)
LPVErrorCode	Save (LString filename, ILRegion *region)
void	SetChannel (int channelIdx, ILImage *channelImage)
void	SetImageData (int w, int h, LPVOID dataPtr, int bpl, BOOL copyData)
void	SetSize (int w, int h)
void	SetTo (int val, ILRegion *region)
void	SetToColor (LColor val, ILRegion *region)
Public Member Functions inherited from ILObject
ILObject *	Copy ()
LPVErrorCode	Load (LString filename)
void	Reset ()
LPVErrorCode	Save (LString filename)
BOOL	Valid ()

Properties
int	BytesPerLine [get]
	The number of bytes contained in a row, include the padding bytes.
LPVOID	DataPointer [get]
	The pointer to the image data buffer. Please DON'T change the content of the data block.
int	Height [get]
	The height of the image, number of rows.
LPVImageFormat	ImageFormat [get, set]
	The format of the image.
int	Pixel (int x, int y) [get, set]
LColor	PixelColor (int x, int y) [get, set]
int	Width [get]
	The width of the image, number of columns.

Detailed Description

This interface expose APIs to access/manipulate/draw a LPV LImage object.

Image is the core object in a vision system. The image class encapsulate all the functionality needed to represent a rectangular shaped image. It stores pixels as a 2-dimensional dense array, row by row.

To use this interface, you should create a LImage object.

There are many different ways to initialize data of the image. The most popular options are listed below:

1. Load image from file via ILImage::Load().
2. Set image data for data pointer via ILImage::SetImageData(), typically you may get this pointer from camera's callback function.
3. Set image size as wanted, then assign pixel values one by one manually. But please noted, this way is not efficient.

Example Code

C++

// 1. load from file
img->Load(L"img.bmp");
// 2. load from data pointer
img->SetImageData(width, height, dataPtr, bytesPerLine, true);
// 3. set pixels manually
img->SetSize(100, 100);
img->PutPixel(10, 20, 255); // set (10, 20) to be a white pixel

C#

// 1. load from file
img.Load("img.bmp");
// 2. load from data pointer
img.SetImageData(width, height, dataPtr, bytesPerLine, true);
// 3. set pixels manually
img.SetSize(100, 100);
img.Pixel[10, 20] = 255; // set (10, 20) to be a white pixel

COM

// 1. load from file
img->Load(L"img.bmp");
// 2. load from data pointer
img->SetImageData(width, height, dataPtr, bytesPerLine, true);
// 3. set pixels manually
img->SetSize(100, 100);
img->Pixel[10][20] = 255; // set (10, 20) to be a white pixel

Member Function Documentation

CopyFrom()

void CopyFrom	(	ILImage *	img,
		ILRegion *	region,
		int	panX,
		int	panY
	)

Copy the content of the input image and cover the current.

Parameters

[in]	img	The input image. If the input image is bigger or outside with the given x/y offset, the redundant part will be ignored.
[in]	region	The input region to crop the input image. Optional, could be null.
[in]	panX	The x-coordinate of the offset
[in]	panY	The y-coordinate of the offset

CopyTo()

void CopyTo

(

ILImage *

dstImg

)

Copy the image object to destination.
If the current image object has its own data buffer, we'll do another clone of data to the destination image object. Otherwise, we just simply assign the data pointer to it.

Parameters

[out]

dstImg

The destination image object.

Draw()

void Draw	(	HDC	hdc,
		double	zoomX,
		double	zoomY,
		double	panX,
		double	panY
	)

Draw the image onto the provided device context.

See also

DrawAsMask()

Use Draw Functions

Remarks

This function is currently not available on Linux / MacOS platform.

DrawAsMask()

void DrawAsMask	(	HDC	hdc,
		double	zoomX,
		double	zoomY,
		double	panX,
		double	panY
	)

Draw the image as a mask onto the provided device context, white pixels preserve the original values on HDC.

See also

Draw()

Use Draw Functions

Remarks

This function is currently not available on Linux / MacOS platform.

GetChannel()

ILImage* GetChannel

(

int

channelIdx

)

Extract the channel of the given index. This function is only for a color image.

Parameters

[in]

channelIdx

The input channel index, 0, 1 or 2.

Return values

channelImage

Return the channel image. It's a grayscale image of the same size as the current color image.

LoadRawFile()

LPVErrorCode LoadRawFile	(	LString	filename,
		int	w,
		int	h,
		int	flags
	)

Load image from .raw file.

Parameters

[in]	w	The width of the image in pixels, should be a positive value. Since .raw format does not include informations for image size, the caller should set the size manually.
[in]	h	The height of the image in pixels, should be a positive value.
[in]	flags	Controls how to parse the raw data Flags Usage 0 Each pixel in 16 bits 1 Two MONO-12 pixels are packed into 24 bits, the low-4 bits of two pixels encapsulated into a 8 bits
[in]	filename	Full path of the file, it could be absolute or relative path with file name and extension.

Return values

error

Return error code if anything is wrong.

See also

Load()

Since

2.7.0

Save()

LPVErrorCode Save	(	LString	filename,
		ILRegion *	region
	)

Save the image object to a specified file. The image format is chosen based on the filename extension. Currently, we support save as BMP, PNG, JPG, JPEG formats.

Parameters

[in]	region	Optional, could be null. If you want to only save part of the image, pass in a region.
[in]	filename	Full path of the file, it could be absolute or relative path with file name and extension.

Return values

error

Return error code if anything is wrong.

See also

Save(), Load()

SetChannel()

void SetChannel	(	int	channelIdx,
		ILImage *	channelImage
	)

Set the channel of the given index to the input grayscale image. This function is only for a color image.

Parameters

[in]	channelIdx	The input channel index, 0, 1 or 2.
[in]	channelImage	The input grayscale channel image, it should be of the same size as the current color image.

SetImageData()

void SetImageData	(	int	w,
		int	h,
		LPVOID	dataPtr,
		int	bpl,
		BOOL	copyData
	)

Assign data to the image object via an external allocated buffer.
If copyData is set to False, this pointer overwrite the internally allocated buffer inside the image object. You should take the responsibility to keep the buffer alive during the entire life cycle of the image object.
If copyData is set to True, we'll copy the content into the buffer owned by the image object. You don't need to worry about the data life cycle then.

Parameters

[in]	w	The width of the image in pixels, should be a positive value.
[in]	h	The height of the image in pixels, should be a positive value.
[in]	dataPtr	Pointer to the external image data buffer.
[in]	bpl	Bytes per line, the number of bytes contained in a row. The value should include the padding bytes. at the end of each row, if any. If set to 0, no padding is assumed and the actual stride is calculated as column number * pixel size(sizeof(char)).
[in]	copyData	Whether to copy the data into owned buffer of the image object.

SetSize()

void SetSize	(	int	w,
		int	h
	)

Change the size of image to provided size. The pixel values are undefined. You may set them all to 0 or other preferred value via SetTo() or SetToColor(). If the image is already of the same, we keep the old data. Pass in 0 width or 0 height to reset the image to be empty.

Parameters

[in]	w	Required image width.
[in]	h	Required image height.

SetTo()

void SetTo	(	int	val,
		ILRegion *	region
	)

Set the pixels in the given region to the specified grayscale value. For a color image, all its channels are set to the same value.

Parameters

[in]	val	The grayscale value
[in]	region	The input region. If the input region is bigger or outside of the image, the redundant part will be ignored. Optional, could be null, which mean the entire image.

SetToColor()

void SetToColor	(	LColor	val,
		ILRegion *	region
	)

Set the pixels in the given region to the specified RGB color value. This function is only for a color image.

Parameters

[in]	val	The color value
[in]	region	The input region. If the input region is bigger or outside of the image, the redundant part will be ignored. Optional, could be null, which mean the entire image.

Property Documentation

Pixel

int Pixel

getset

The pixel's grayscale value.

Parameters

[in]	x	The x-coordinate of the pixel.
[in]	y	The y-coordinate of the pixel.

PixelColor

LColor PixelColor

getset

The pixel's color value.

Parameters

[in]	x	The x-coordinate of the pixel.
[in]	y	The y-coordinate of the pixel.