Countious binary output of a single column

Beginners
1

background - i’ve purchased a commercial lic to get the higher resolution, my files are 4k 120fps (though a solution should work fine with free lic.)

what i’m trying to create : a X*Y file where resy =y, true_length =X of the video in
not opening a image for each frame but contentiously saving
one column frame of data to the same file -

I.E create a long slitscan image from a mov file

  • once in raw byte format i can use raw2tiff to put a proper header on it (tiff file should be ~1-2Gbyte)

i’m currently at a point where i’m getting the column, is it possible just to save the numpy.array() from a TOP or do i have to go though a CHOP …

2

Hi @ralphnev,

not sure how performant that is but you could do this in a Script TOP:

# me - this DAT
# scriptOp - the OP which is cooking

import numpy

# press 'Setup Parameters' in the OP to call this function to re-create the parameters.
def onSetupParameters(scriptOp):
	page = scriptOp.appendCustomPage('Column Recorder')
	p = page.appendToggle('Record', label='Record')
	return

# called whenever custom pulse parameter is pushed
def onPulse(par):
	return

record = 0
frameCount = 0
image = numpy.empty((1,1,4), dtype=numpy.float32)

def onCook(scriptOp):
	global record, frameCount, image
	# fetch the input as a numpy array


	if record == 0 and scriptOp.par.Record == 1:
		record = 1
		frameCount = 0
		input = scriptOp.inputs[0].numpyArray(delayed=True)
		image = numpy.array(input)
	elif record == 1 and scriptOp.par.Record == 1:
		frameCount += 1
		# append the input as a new column to the image array
		input = scriptOp.inputs[0].numpyArray(delayed=True)
		image = numpy.append(image, input, axis=1)
	elif record == 1 and scriptOp.par.Record == 0:
		record = 0
		frameCount = 0
		# save the image to a file
		
		# # # # # # your code here
			
		# reset image to empty
		image = numpy.empty((1,1,4), dtype=numpy.float32)

	# copy it into the top
	scriptOp.copyNumpyArray(image)
	return

If you think you might exceed the size possible for the videocard’s limit, you could skip the display as well…

cheers
Markus

3

sorry for the delay, but yes, - edit : figured it out … still needs some finessing but it works :wink:
IE.

                # # # # # # your code here
		image = numpy.delete(image,3,2) #delete alpha layer
		image *= 256 # or any coefficient 
		image2 = image2.astype(numpy.uint8)

		scriptOp.copyNumpyArray(image)
		scriptOp.save('c:/TEMP/long.tiff')