Description
LumiScan Area Detector finds regions in a camera image and reports how big each one is.
Unlike the other LumiScan matching applications it needs no reference picture. You describe what counts as a region - how bright, how large, how round - and it reports every region matching that description, with its area and position.
That suits counting parts, checking coverage or contamination, measuring a fill level, and any inspection where the thing being measured has no fixed appearance: a spill, a scratch, a gap.
|
If you are looking for a specific known part, the sibling applications are the better fit: LumiScan Template Finder matches a stored picture at a range of scales and rotations, and LumiScan Object Finder matches an object’s shape. Area Detector answers "what is here, and how big", not "where is this thing". |
How a region gets found
Every method measures - the reported area is always a pixel count. What input/method chooses is how
a region is found in the first place:
input/method is FIXED: the image is thresholded once at a level you commission, and every
connected region is measured by counting its pixels, whatever its shape. It is the only method this
version offers, and anything else written to that node is refused rather than ignored.
A region is counted as the pixels that passed the threshold, so a hole inside a part is not counted
unless fill-holes is set — which is a choice about what a washer’s area means, not a
correction.
To find the level, run Analysing a Scene: it measures the picture, suggests a level, and publishes the grey-level histogram that level should sit in the valley of.
Where the images come from
The application does not talk to a camera itself. It reads an image node published by another application on the Data Layer, given as an address, for example:
hdvisionsystems/lumiscan-cam-driver/camera1/output/image
The node must publish data, width, height and pixel-type, which is what LumiScan Cam Driver
provides.
Modules
A module is one configured instance of the detection: its own camera address, its own settings and its own results, at its own Data Layer address. One installation can run several modules to watch several cameras at once, and one license covers ten of them. See Module Management.
Modules share nothing: there is no stored reference image, so there is nothing application-wide to configure.