Using undefined image data

You can open undefined or raw image data the same way you would any other file.

To open a raw or undefined image

In File Selector, select the raw image you want, and then click Open.
The Raw Imagery File Definition window appears, in which you can define the format of the raw imagery file.
Under Raw Imagery File Definition, in the Header Bytes box, type the number of bytes to allocate for header information.
The zero value indicates that imagery data starts at the beginning of the file.
Under File Dimensions, do the following:
- Type the size of the image in the Pixels and Lines boxes, respectively.
  Pixels are the x-value and lines are the y-value. The values should be the full size of the image in the file, even if you intend only to open a subset of the file.
- In the Channels box, type the number of channels in the file.
Beside Data Interleaving, select how multiple channels of image data are interleaved.
- Pixel: The channels are pixel interleaved. For example, in a three-channel file, the values in the file are 123123123..., with the channel values for a given pixel located together.
- Line: The channels are line interleaved. The data for line 1 of the first channel occurs first, followed by the data for line 1 of the second channel and line 1 of the third channel. Next is line 2 of the first channel, and so on. For example, in a three-channel file, the values in the file would be (line 1) 111...222...333, (line 2) 111...222...333, and so on.
- Band: The channels are band sequential. All the data for the entire first channel would be first, followed by all the data for the second channel, and so on.
Note: Data Interleaving is not applicable to one-channel images. Single-band data can have band interleaving.
Beside Data type, click to define the type of data:
- 8-bit: Data of each channel is 8-bit
- 16-bit unsigned: Data of each channel is 16-bit, unsigned. Values range from 0 to 65535, and are two bytes each.
- 16-bit signed: Data for each channel is 16-bit, signed. Values range from -32768 to 32767, and are two bytes each.
- 32-bit real: Each pixel is stored as four bytes of data, and can have an IEEE floating-point value ranging from -3.4E8 to 3.4E8.
- 16-bit signed complex
- 32-bit real complex
Beside Byte Order, select how to store the order of multiple-byte data words:
- MSB: Most significant byte first.
  This is the order common on Sun, IBM, RS/6000, HP, SGI, and Mac systems. It is also known as unswapped, big endian, or Motorola order. With MSB, a 16U-pixel value of one would be expressed as two bytes, the first a zero, and the second a one. The default selected on the panel is the byte order of the local system.
- LSB: Least-significant byte first.
  This order is common on IBM computers with Intel 80 x 86 architecture. It is sometimes known as swapped or little endian. With LSB, a 16U-pixel value of 1 would be expressed as two bytes, the first a 1, and the second a 0.
Note: Byte order is for non-8-bit image data only; it can be ignored for files containing only 8-bit data.
Click Accept.

A new header file is created with a .pox file name extension. It is an auxiliary file that contains layout information for the imagery.

The following is an example of the file produced for a simple 1000 x 1000 8-bit single-channel image.

Auxiliary Target: rawimage.bil
Raw Definition: 1000 1000 1
ChanDefinition -1: 8U 0 1 1000 Swapped

Note: After the configuration file (.pox) is created, you cannot redefine the raw file until you delete the .pox file. When experimenting with possible raw format definitions, using a .pox file is not recommended.