Improvement for prom image

[mconsidine]

Improvement for prom image

[Vdesnoux]

Hi Matt
I added a disk in the new version, and fixed a bug on the thresholding. Lett's see if it meets the expectations...

[mconsidine]

Hi,
I'm attaching here a comparison of the the prom results, one being what I think is the current way in Solex_ser_recon and the other reflects the adjustment I suggested above. "16_33_15_pipp_protus.png" is the new version ...

The relevant portion of code that generated this is
`
#prom image thresholds

if disk_display==True:
    x0=circle[0]
    y0=circle[1]
    wi=circle[2]-5
    he=circle[3]-5
    #ang=-int(circle[3])
    r=int(min(wi,he)-3)
    c=(0,0,0)
    #r=int(circle[2])-5
    #frame_contrasted3=cv2.circle(frame_contrasted3, (x0,y0),r,80,-1,lineType=cv2.LINE_AA)
    frame=cv2.ellipse(frame, (x0,y0),(wi,he),-circle[4],0,360,(0,0,0),-1,lineType=cv2.LINE_AA ) #MattC apply tilt, change color to black
frame1=np.copy(frame)
Threshold_Upper=np.percentile(frame1,99.9999) #MattC drop *.18
Threshold_low=0
#print('Threshold Lower prom', Threshold_low)
#print('Threshold Upper prom', Threshold_Upper)
frame1[frame1>Threshold_Upper]=Threshold_Upper
fc2=(frame1-Threshold_low)* (scalemax/(Threshold_Upper-Threshold_low))
fc2[fc2<0]=0
frame_contrasted3=np.array(fc2, dtype=datasize)
cv2.imshow('proms',frame_contrasted3)

`

(Apologies for supply an English-language version. But it should be easy to see that I just moved one block of code further up and made a couple of small adjustments reflected by my initials).

To my eye, the result seems better when I compare them side by side in GIMP, for instance.

I hope this helps!

[Vdesnoux]

I modified the code as per your suggestion, but sill keep some threshold high. So, done !