TI Thinks Resolved

OPT8241-CDK-EVM: getting a snapshot that showing an R(t) curve for the image

Genius 5235 points

Shai Berman

Replies: 9

Views: 324

Part Number: OPT8241-CDK-EVM

Hi team,

my customer owns a kit and he would like to know the following:

If possible, I'd like to be able to get a snapshot showing an R(t) curve for the image, that is, what is the reflectance [R] intensity as a function of time [t].

The application I wish to explore is diffusive and based on past studies should have that kind of behavior.

As an analogy, consider imaging a thick glass slab, there will be some reflectance from the front air/glass interface and a bit later there will be a reflectance from the rear glass/air interface. What I'm most interested in is knowing the temporal value of that reflectance.

I know this is not what the system is designed for but would like to know if it's possible.

can you please advise the above, thanks in advance.

Best regards,

Shai

Genius 5235 points

Shai Berman
- Oct 31, 2018 4:00 PM
Hi team, there is an update?
Thanks in advance,
Shai
- Locked
- Cancel
Do you have another question?

We are glad that we were able to resolve this issue, and will now proceed to close this thread.

If you have further questions related to this thread, you may click "Ask a related question" below. The newly created question will be automatically linked to this question.

Ask a related question Ask a new question
Intellectual 1610 points

Nithin Jose
- Nov 6, 2018 3:29 AM
In reply to Shai Berman:
Hi,

Can you help us in understanding your application slightly better. Are you looking at a stationary object ? Is the target a lambertian surface ?

Regards
Nithin
- Locked
- Cancel
Do you have another question?

We are glad that we were able to resolve this issue, and will now proceed to close this thread.

If you have further questions related to this thread, you may click "Ask a related question" below. The newly created question will be automatically linked to this question.

Ask a related question Ask a new question
Genius 5235 points

Shai Berman
- Nov 10, 2018 7:07 PM
In reply to Nithin Jose:

Nithin,

please find my customer comments inline:

Can you help us in understanding your application slightly better? - We would like to use the camera for a bio-medical application, i.e. tissue imaging.

We were able to get some very interesting results for the camera as is with slight augmentations but are interested in some specific not so standard measurement.

Are you looking at a stationary object? - Yes, the distance to the target is fixed and as near as we can while avoiding saturation.

Is the target a Lambertian surface? - No, we are imaging a tissue which is somewhat permeative in the 850nm range (i.e. partly transparent)

Thanks in advance

BR

Shai
- Locked
- Cancel
Do you have another question?

We are glad that we were able to resolve this issue, and will now proceed to close this thread.

If you have further questions related to this thread, you may click "Ask a related question" below. The newly created question will be automatically linked to this question.

Ask a related question Ask a new question
Intellectual 1610 points

Nithin Jose
- Nov 14, 2018 3:41 AM
In reply to Shai Berman:

Shai,

This can possibly be done. However TI's expertise does not lie in the corresponding field. You might have to refer to some academic material for your application.

On a side note, the amplitude shown return by the camera is a measure of the returned signal. Reflectivity of the target is one of the parameters that affect this value.

Regards

Nithin
- Locked
- Cancel
Do you have another question?

We are glad that we were able to resolve this issue, and will now proceed to close this thread.

If you have further questions related to this thread, you may click "Ask a related question" below. The newly created question will be automatically linked to this question.

Ask a related question Ask a new question
Genius 5235 points

Shai Berman
- Nov 15, 2018 10:54 AM
In reply to Nithin Jose:
Nithin,

Thank you for this reply, the customer has actually been able to show some potential in that direction but he would like to know how to get closer to the "raw" signal.
Is there a way to get the time-dependent reflection function for a pixel, before it is being reduced to the 3D image.
The customer has the academic knowledge to manage the results, but not the tech know how to extract that from the ICs registers.
Could you please advise, Thanks.

Regards,
Shai
- Locked
- Cancel
Do you have another question?

We are glad that we were able to resolve this issue, and will now proceed to close this thread.

If you have further questions related to this thread, you may click "Ask a related question" below. The newly created question will be automatically linked to this question.

Ask a related question Ask a new question
Genius 5235 points

Shai Berman
- Nov 25, 2018 12:13 PM
In reply to Shai Berman:
Hello Team,

To clarify a bit my question, it is possible to extract/capture the 'RAW' data that comes from OPT8241 ADC in any way?
Thanks in advance.

Regards,
Shai
- Locked
- Cancel
Do you have another question?

We are glad that we were able to resolve this issue, and will now proceed to close this thread.

If you have further questions related to this thread, you may click "Ask a related question" below. The newly created question will be automatically linked to this question.

Ask a related question Ask a new question
Expert 3780 points

Subhash Chandra Sadhu
- Nov 27, 2018 4:59 AM
In reply to Shai Berman:
Shai,

The TFC (OPT9221) typically processes all the data that comes from the ADC(OPT8241). However, we are able to give a lot of flexibility on how TFC actually performs the computation.
Customer can set the camera in a single frequency 6 quad mode. In this mode, they can program whatever coefficients they would like to and the TFC will give appropriate results. It is not directly possible to get the raw data, but they can get data from which the raw data can be computed.
Calculations inside the ToF controller:

A = I + iQ
Q0 = |A|cos (theta)
Q1 = |A|cos(theta-60)

Q = [Q0 , Q1, Q2, .... , Q5] ## This vector is the analog voltage output of the pixel
C = [[c0,c1,c2,c3,c4,c5],[s0,s1,s2,s3,s4,s5]] ##This matrix is the coefficient matrix which gets multiplied with digitized Q
[[I],[Q]] = CT*QT ##This is the multiplication operation performed by OPT9221
Amplitude = sqrt(I*I+Q*Q)
Phase = atan(Q/I)
Q = Amplitude * sin(phase)
I = Amplitude *cos(phase)

The TFC gives "phase" and "amplitude" data only. But given that these are the computations performed by the TFC, it is possible to change all the values of the coefficient matrix, C to achieve what they want.

Regards,
Subhash

3D ToF Product Landing Page
- Locked
- Cancel
Do you have another question?

We are glad that we were able to resolve this issue, and will now proceed to close this thread.

If you have further questions related to this thread, you may click "Ask a related question" below. The newly created question will be automatically linked to this question.

Ask a related question Ask a new question
Genius 5235 points

Shai Berman
- Nov 29, 2018 10:05 AM
In reply to Subhash Chandra Sadhu:
Subhash,

Thank you for your input, I looked again in the documents and white papers in the TI website and didn't find any referencing the equations mentioned here.
I think that getting the Q vector per pixel is what the customer were asking for, could you please advise how to retrieving that? Where are the explanations on how to use any pre-made SW, or of any required FW changes?

Thanks in advance.

Regards,
Shai
- Locked
- Cancel
Do you have another question?

We are glad that we were able to resolve this issue, and will now proceed to close this thread.

If you have further questions related to this thread, you may click "Ask a related question" below. The newly created question will be automatically linked to this question.

Ask a related question Ask a new question
Expert 3780 points

Subhash Chandra Sadhu
- Nov 29, 2018 10:17 AM
In reply to Shai Berman:

Shai,

This is a capability of OPT9221. Customer can program any value for the Q vector. The actual values are described in Equation 19 in the datasheet.
The values suggested in the datasheet are the values which we recommend. However, customer can program their own values to any other values and get information which is representative of the raw data.

Regards,
Subhash

3D ToF Product Landing Page
- Locked
- Cancel
Do you have another question?

We are glad that we were able to resolve this issue, and will now proceed to close this thread.

If you have further questions related to this thread, you may click "Ask a related question" below. The newly created question will be automatically linked to this question.

Ask a related question Ask a new question