VPUTCAL (camera) scalers[i], pmm.to.pix[j,k],
pix.to.pmm[l,m], to.cam

Function

Load vision calibration data from 1 to 3 arrays.

Usage Considerations

The VISION switch must be enabled, the vision processor must be idle, and vision model training must not be active for this instruction to be executed.

Parameter s

camera Real-valued expression that specifies the virtual camera.

scalers[] Real array of calibration values (see below).

pmm.to.pix[,] Optional, two-dimensional real array that defines the millimeter-to-pixel transformation matrix. If this array is specified, the array pix.to.pmm[,] must also be specified.

pix.to.pmm[,] Optional, two-dimensional real array that defines the pixel-to-millimeter transformation matrix. If this array is specified, the array pmm.to.pix[,] must also be specified.

i, j, k, l, m Optional integer values that identify the first array element to be referenced in the respective array. Zero is assumed for each index that is omitted. If an array is specified that has more dimensions than needed, only the rightmost indexes are incremented as the values are referenced.

to.cam Optional transformation expression that defines the vision transformation to store with the calibration data. (Note: in V+ version 11.0, to.cam must be a variable. In version 11.1 and higher, to.cam does not have to be a variable.)

Details

The calibration programs supplied with the AdeptVision systems fill the calibration arrays and execute a VPUTCAL instruction at the end of a successful calibration. This calibration data is used when performing operations that involve millimeter distances or locations in the image, such as VPICTURE, VWINDOW, VRULER, VLOCATE, and VFEATURE.

Since the calibration arrays are generally filled by an Adept utility program, the programmer should not have to be familiar with all the individual array elements. However, since some elements may be useful to know about, a brief description of the array elements is provided in Elements of VGETCAL/VPUTCAL Scaler Calibration Array.

Any undefined elements of the input arrays are automatically assumed to be zero. Subsequent VGETCAL instructions will retrieve arrays that are defined completely.

NOTE: The user is discouraged from setting any of the array values unless explicitly requested to do so by Adept documentation or personnel. The descriptions here are intentionally terse.

Several of the entries in the scaler array documented here are not actually used by the vision system but are generated by the Adept calibration utility programs and used by other Adept software. Those entries, identified here only for completeness, are:

Element 2: Calibration method
Element 3: Link number
Element 13, bit 1: HPS-used flag

The value for xy.ratio must be in the range 0.5 to 1.5. If the x.scale/y.scale supplied to VPUTCAL is not in the acceptable range, the error

	*Invalid vision X/Y ratio*  

is reported when the next VPICTURE operation is processed.

The arrays pmm.to.pix[,] and pix.to.pmm[,] contain two-dimensional homogeneous transformations. These transformations implicitly contain the x and y millimeter-per-pixel scaling contained in the array scaler[]. In addition, the transformations may contain terms that correct for perspective distortion. If the two transformations are not specified with VPUTCAL, correction for perspective distortion is automatically disabled. (See Appendix C of the AdeptVision Reference Guide for more information on perspective distortion.)

If quick-planning is used, you may experience slower average and worst-case recognition times. Therefore, you should do a full planning before proceeding with normal runtime activities.

If the new weights after a learning mode VFINDER do not result in a new ordering of the feature classes, the learning convergence count is incremented. If there is a new order, this count is set to 0. Therefore, if there is no change in the order of the feature classes for several scenes in a row, this count will rise and can be used to indicate that you can exit the learning mode.

Example

Calibrate virtual camera 2 using the data in cal.array[ ]:

	VPUTCAL (2) cal.array[] 	 

Related Keyword

VGETCAL (cam) scalers[i], pmm.to.pix[j,k], pix.to.pmm[l,m], to.cam (program instruction)

Syntax

VTRAIN.FINDER-I