How laser rangefinder or laser distance meter work and what chip can measure Time of Flight ?...

[Michael Kellett]

On 13/08/2022 18:42, Phil Hobbs wrote:

The usual method is a time-to-amplitude converter.

Cheers

Phil Hobbs

TI make some nice chips, you can even buy them:

TDC7201, TDC7200

Resolution: 55 ps
• Standard Deviation: 35 ps
• Measurement Range:
– Mode 1: 12 ns to 500 ns
– Mode 2: 250 ns to 8 ms

It needs a 16MHz input clock, work it out before you look at the data sheet.

MK

 

[server]

On Sat, 13 Aug 2022 13:42:30 -0400, Phil Hobbs
<pcdhSpamMeSenseless@electrooptical.net> wrote:

Ed Lee wrote:
On Friday, August 12, 2022 at 3:40:38 PM UTC-7, a a wrote:
On Friday, 12 August 2022 at 23:58:11 UTC+2, Ed Lee wrote:
On Friday, August 12, 2022 at 2:31:31 PM UTC-7, a a wrote:
On Friday, 12 August 2022 at 22:37:28 UTC+2, Ed Lee wrote:
On Friday, August 12, 2022 at 1:06:35 PM UTC-7, a a wrote:
On Friday, 12 August 2022 at 21:55:26 UTC+2, Ed Lee wrote:
On Friday, August 12, 2022 at 11:45:18 AM UTC-7, a a wrote:
On Friday, 12 August 2022 at 15:46:11 UTC+2, Ed Lee wrote:
On Friday, August 12, 2022 at 3:23:57 AM UTC-7, a a wrote:
On Friday, 12 August 2022 at 12:18:44 UTC+2, a a wrote:
what chip, clocked at what frequency, is used to measure Time of Flight of laser light at such short time intervals ?
GHz ASIC
3E-9 second per meter
c = 299 792 458 m/s > >>> c = 300 000 000 m/s

3E-9 s/m x 300 000 000 m/s = 0.9

ok
but for 1 cm resolution
we need 100 x faster clock

and for parallel analysis of point cloud 100 x 100
we need 100 x 100 faster clock

could you explain ?
There is no point in measuring all 10,000 points all at once.
so do you suggest,
what is marketed by iPhone and called Lidar in smartphone,
is a single point Laser Range Meter functionality ?

If you are correct, so why do they present 2D laser scanner functionality on images ?

Single point laser Lidar requires rotating head to work

\\so some kind od mechanics is involved
Yes, precisely, to have better relative positioning with multi-sensors. However, there is no need to measure them all within E-9 second.
-E-9 second cloc k is for 1m resolution
for 1 cm resolution
you need
E-9 * 10-2 second clock

but we still discuss a single-point operation
With 5GHz (close to current fab cap) hardware counters, we can count pulses to the same point 100 times. If we get 25 pulses of 1 meter and 75 pulses of 1.1 meter, we can guess that the distance is close to 1.025 meters.

Give me $100,000 and i can build you the chip to prove it.
-==c = 299 792 458 m/s > >>> c = 300 000 000 m/s

-===3E-9 s/m x 300 000 000 m/s = 0.9

300 000 000 m/s = 100 * 300 000 000 cm/s

= 30 000 000 000 cm/s

so you need 30 GHz clock x 2 to get 1 cm resolution for a single point

3GHz (reasonable clock) for 10cm resolution.

ok, for 1 m distant object
300 MHz x 2 clock can do the job

but if you need mobile laser range meter to scan objects on the fly, to act as 2D scanner,
1 cm counts and makes the difference

https://en.wikipedia.org/wiki/Oversampling
for higher resolution.


The usual method is a time-to-amplitude converter.

Cheers

Phil Hobbs

One way is to use a simple analog time stretcher and a
modest-frequency counter.

 

[a a]

On Sunday, 14 August 2022 at 18:24:59 UTC+2, jla...@highlandsniptechnology.com wrote:

On Sat, 13 Aug 2022 13:42:30 -0400, Phil Hobbs
pcdhSpamM...@electrooptical.net> wrote:

Ed Lee wrote:
On Friday, August 12, 2022 at 3:40:38 PM UTC-7, a a wrote:
On Friday, 12 August 2022 at 23:58:11 UTC+2, Ed Lee wrote:
On Friday, August 12, 2022 at 2:31:31 PM UTC-7, a a wrote:
On Friday, 12 August 2022 at 22:37:28 UTC+2, Ed Lee wrote:
On Friday, August 12, 2022 at 1:06:35 PM UTC-7, a a wrote:
On Friday, 12 August 2022 at 21:55:26 UTC+2, Ed Lee wrote:
On Friday, August 12, 2022 at 11:45:18 AM UTC-7, a a wrote:
On Friday, 12 August 2022 at 15:46:11 UTC+2, Ed Lee wrote:
On Friday, August 12, 2022 at 3:23:57 AM UTC-7, a a wrote:
On Friday, 12 August 2022 at 12:18:44 UTC+2, a a wrote:
what chip, clocked at what frequency, is used to measure Time of Flight of laser light at such short time intervals ?
GHz ASIC
3E-9 second per meter
c = 299 792 458 m/s > >>> c = 300 000 000 m/s

3E-9 s/m x 300 000 000 m/s = 0.9

ok
but for 1 cm resolution
we need 100 x faster clock

and for parallel analysis of point cloud 100 x 100
we need 100 x 100 faster clock

could you explain ?
There is no point in measuring all 10,000 points all at once.
so do you suggest,
what is marketed by iPhone and called Lidar in smartphone,
is a single point Laser Range Meter functionality ?

If you are correct, so why do they present 2D laser scanner functionality on images ?

Single point laser Lidar requires rotating head to work

\\so some kind od mechanics is involved
Yes, precisely, to have better relative positioning with multi-sensors. However, there is no need to measure them all within E-9 second.
-E-9 second cloc k is for 1m resolution
for 1 cm resolution
you need
E-9 * 10-2 second clock

but we still discuss a single-point operation
With 5GHz (close to current fab cap) hardware counters, we can count pulses to the same point 100 times. If we get 25 pulses of 1 meter and 75 pulses of 1.1 meter, we can guess that the distance is close to 1.025 meters.

Give me $100,000 and i can build you the chip to prove it.
-==c = 299 792 458 m/s > >>> c = 300 000 000 m/s

-===3E-9 s/m x 300 000 000 m/s = 0.9

300 000 000 m/s = 100 * 300 000 000 cm/s

= 30 000 000 000 cm/s

so you need 30 GHz clock x 2 to get 1 cm resolution for a single point

3GHz (reasonable clock) for 10cm resolution.

ok, for 1 m distant object
300 MHz x 2 clock can do the job

but if you need mobile laser range meter to scan objects on the fly, to act as 2D scanner,
1 cm counts and makes the difference

https://en.wikipedia.org/wiki/Oversampling
for higher resolution.


The usual method is a time-to-amplitude converter.

Cheers

Phil Hobbs
One way is to use a simple analog time stretcher and a
modest-frequency counter.
just watched video

kinect - sensor IR projection

https://youtu.be/MlTf0yYQjSg?t=43

let me know if the projected IR points cloud is made of static preset points
and what is a role of the pattern ?

Can theory behing Kinect be turned into large scale indoor 3D scanner,
replacing lenses in ca,meras,
moving IR projector far from VGA camera to read depth at large distance ?

Can IR pattern projected be replaced by another, higher IR power/ higher resolution pattern ?

 

[Ed Lee]

On Sunday, August 14, 2022 at 1:56:01 PM UTC-7, a a wrote:

On Sunday, 14 August 2022 at 18:24:59 UTC+2, jla...@highlandsniptechnology.com wrote:
On Sat, 13 Aug 2022 13:42:30 -0400, Phil Hobbs
pcdhSpamM...@electrooptical.net> wrote:

Ed Lee wrote:
On Friday, August 12, 2022 at 3:40:38 PM UTC-7, a a wrote:
On Friday, 12 August 2022 at 23:58:11 UTC+2, Ed Lee wrote:
On Friday, August 12, 2022 at 2:31:31 PM UTC-7, a a wrote:
On Friday, 12 August 2022 at 22:37:28 UTC+2, Ed Lee wrote:
On Friday, August 12, 2022 at 1:06:35 PM UTC-7, a a wrote:
On Friday, 12 August 2022 at 21:55:26 UTC+2, Ed Lee wrote:
On Friday, August 12, 2022 at 11:45:18 AM UTC-7, a a wrote:
On Friday, 12 August 2022 at 15:46:11 UTC+2, Ed Lee wrote:
On Friday, August 12, 2022 at 3:23:57 AM UTC-7, a a wrote:
On Friday, 12 August 2022 at 12:18:44 UTC+2, a a wrote:
what chip, clocked at what frequency, is used to measure Time of Flight of laser light at such short time intervals ?
GHz ASIC
3E-9 second per meter
c = 299 792 458 m/s > >>> c = 300 000 000 m/s

3E-9 s/m x 300 000 000 m/s = 0.9

ok
but for 1 cm resolution
we need 100 x faster clock

and for parallel analysis of point cloud 100 x 100
we need 100 x 100 faster clock

could you explain ?
There is no point in measuring all 10,000 points all at once.
so do you suggest,
what is marketed by iPhone and called Lidar in smartphone,
is a single point Laser Range Meter functionality ?

If you are correct, so why do they present 2D laser scanner functionality on images ?

Single point laser Lidar requires rotating head to work

\\so some kind od mechanics is involved
Yes, precisely, to have better relative positioning with multi-sensors. However, there is no need to measure them all within E-9 second.
-E-9 second cloc k is for 1m resolution
for 1 cm resolution
you need
E-9 * 10-2 second clock

but we still discuss a single-point operation
With 5GHz (close to current fab cap) hardware counters, we can count pulses to the same point 100 times. If we get 25 pulses of 1 meter and 75 pulses of 1.1 meter, we can guess that the distance is close to 1.025 meters.

Give me $100,000 and i can build you the chip to prove it.
-==c = 299 792 458 m/s > >>> c = 300 000 000 m/s

-===3E-9 s/m x 300 000 000 m/s = 0.9

300 000 000 m/s = 100 * 300 000 000 cm/s

= 30 000 000 000 cm/s

so you need 30 GHz clock x 2 to get 1 cm resolution for a single point

3GHz (reasonable clock) for 10cm resolution.

ok, for 1 m distant object
300 MHz x 2 clock can do the job

but if you need mobile laser range meter to scan objects on the fly, to act as 2D scanner,
1 cm counts and makes the difference

https://en.wikipedia.org/wiki/Oversampling
for higher resolution.


The usual method is a time-to-amplitude converter.

Cheers

Phil Hobbs
One way is to use a simple analog time stretcher and a
modest-frequency counter.
just watched video

kinect - sensor IR projection

https://youtu.be/MlTf0yYQjSg?t=43

let me know if the projected IR points cloud is made of static preset points
and what is a role of the pattern ?

Can theory behing Kinect be turned into large scale indoor 3D scanner,
replacing lenses in ca,meras,
moving IR projector far from VGA camera to read depth at large distance ?

Can IR pattern projected be replaced by another, higher IR power/ higher resolution pattern ?

Nobody\'s talking, so i will.

This IR stuff has nothing to do with measuring distance, unless you have to do it in the dark. If you are asking about 3D distance mapping, you have to enable it one at a time, even if you have a 100x100 laser matrix.

As someone pointed out, you can buy existing chips. But if you are integrating it into laser matrix, you might have to build your own hardware counters and routing circuits. 5GHz counters can handle at least an inch or couple cm resolution, before oversampling. You can turn off most of the circuit except for fraction of a second, in order to save power.

 

[a a]

On Monday, 15 August 2022 at 14:51:32 UTC+2, Ed Lee wrote:

On Sunday, August 14, 2022 at 1:56:01 PM UTC-7, a a wrote:
On Sunday, 14 August 2022 at 18:24:59 UTC+2, jla...@highlandsniptechnology.com wrote:
On Sat, 13 Aug 2022 13:42:30 -0400, Phil Hobbs
pcdhSpamM...@electrooptical.net> wrote:

Ed Lee wrote:
On Friday, August 12, 2022 at 3:40:38 PM UTC-7, a a wrote:
On Friday, 12 August 2022 at 23:58:11 UTC+2, Ed Lee wrote:
On Friday, August 12, 2022 at 2:31:31 PM UTC-7, a a wrote:
On Friday, 12 August 2022 at 22:37:28 UTC+2, Ed Lee wrote:
On Friday, August 12, 2022 at 1:06:35 PM UTC-7, a a wrote:
On Friday, 12 August 2022 at 21:55:26 UTC+2, Ed Lee wrote:
On Friday, August 12, 2022 at 11:45:18 AM UTC-7, a a wrote:
On Friday, 12 August 2022 at 15:46:11 UTC+2, Ed Lee wrote:
On Friday, August 12, 2022 at 3:23:57 AM UTC-7, a a wrote:
On Friday, 12 August 2022 at 12:18:44 UTC+2, a a wrote:
what chip, clocked at what frequency, is used to measure Time of Flight of laser light at such short time intervals ?
GHz ASIC
3E-9 second per meter
c = 299 792 458 m/s > >>> c = 300 000 000 m/s

3E-9 s/m x 300 000 000 m/s = 0.9

ok
but for 1 cm resolution
we need 100 x faster clock

and for parallel analysis of point cloud 100 x 100
we need 100 x 100 faster clock

could you explain ?
There is no point in measuring all 10,000 points all at once..
so do you suggest,
what is marketed by iPhone and called Lidar in smartphone,
is a single point Laser Range Meter functionality ?

If you are correct, so why do they present 2D laser scanner functionality on images ?

Single point laser Lidar requires rotating head to work

\\so some kind od mechanics is involved
Yes, precisely, to have better relative positioning with multi-sensors. However, there is no need to measure them all within E-9 second.
-E-9 second cloc k is for 1m resolution
for 1 cm resolution
you need
E-9 * 10-2 second clock

but we still discuss a single-point operation
With 5GHz (close to current fab cap) hardware counters, we can count pulses to the same point 100 times. If we get 25 pulses of 1 meter and 75 pulses of 1.1 meter, we can guess that the distance is close to 1.025 meters.

Give me $100,000 and i can build you the chip to prove it.
-==c = 299 792 458 m/s > >>> c = 300 000 000 m/s

-===3E-9 s/m x 300 000 000 m/s = 0.9

300 000 000 m/s = 100 * 300 000 000 cm/s

= 30 000 000 000 cm/s

so you need 30 GHz clock x 2 to get 1 cm resolution for a single point

3GHz (reasonable clock) for 10cm resolution.

ok, for 1 m distant object
300 MHz x 2 clock can do the job

but if you need mobile laser range meter to scan objects on the fly, to act as 2D scanner,
1 cm counts and makes the difference

https://en.wikipedia.org/wiki/Oversampling
for higher resolution.


The usual method is a time-to-amplitude converter.

Cheers

Phil Hobbs
One way is to use a simple analog time stretcher and a
modest-frequency counter.
just watched video

kinect - sensor IR projection

https://youtu.be/MlTf0yYQjSg?t=43

let me know if the projected IR points cloud is made of static preset points
and what is a role of the pattern ?

Can theory behing Kinect be turned into large scale indoor 3D scanner,
replacing lenses in ca,meras,
moving IR projector far from VGA camera to read depth at large distance ?

Can IR pattern projected be replaced by another, higher IR power/ higher resolution pattern ?
Nobody\'s talking, so i will.

This IR stuff has nothing to do with measuring distance, unless you have to do it in the dark. If you are asking about 3D distance mapping, you have to enable it one at a time, even if you have a 100x100 laser matrix.

As someone pointed out, you can buy existing chips. But if you are integrating it into laser matrix, you might have to build your own hardware counters and routing circuits. 5GHz counters can handle at least an inch or couple cm resolution, before oversampling. You can turn off most of the circuit except for fraction of a second, in order to save power.

thank you
I need to learn more about laser IR projector in Kinect

\"The Kinect infrared sensor sees the sofa as a large number of tiny dots. The Kinect sensor constantly projects these dots over the area in its view. If you want to view the dots yourself, it’s actually very easy; all you need is a video camera or camcorder that has a night vision mode. A camera in night vision mode is sensitive to the infrared light spectrum that the Kinect distance sensor uses.

Figure 1-6, for example, was taken in complete darkness, with the sofa lit only by the Kinect. The infrared sensor in the Kinect is fitted with a filter that keeps out ordinary light, which is how it can see just the infrared dots, even in a brightly lit room. The dots are arranged in a pseudo-random pattern that is hardwired into the sensor. You can see some of the pattern in Figure 1-7.


https://www.microsoftpressstore.com/articles/article.aspx?p=2201646

I need to know how pseudo-random pattern of dots is generated by IR laser projector
Resolution is not low

https://www.microsoftpressstore.com/content/images/chap1_9780735663961/elementLinks/httpatomoreillycomsourcemspimages1239382.jpg

dots are not lined up so it seems to me some laser optics / lens is involved

- single laser diode + lens with drilled pattern ?

\"pseudo-random pattern that is hardwired into the sensor.

not sure what they mean, since to have pseudo-random[pattern hardwired into the sensor
you need to get (x,y) coordinates for every single point

I recall another algorithm to generate 3D depth images by moving camera, based on blur effect,
since closer objects move faster in the plane perpendicular to the camera axis
so if we stack a number of images/frames together, we get blur effect for the closer objects to intensify.

----
Kinect`s infrared projector in action
51,995 views
17 Nov 2010
37
Dislike
Share
Clip
Save
Vladimir Seregin
374 subscribers
Video recorded with an regular webcam without IR filter on it.
https://www.youtube.com/watch?v=brnIty7mh2Q

since every single dot is clearly visible in regular HD webcam
so a number of dots projected must be below HD resolution

--
there is a number of web links but they refer to IR camera resolution only

640 x 480
The Kinect sensor returns 16 bits perpixel infrared data with a resolution of 640 x 480as an color image format, and it supports up to30 FPS. Following are the couple of images ( taken in a complete dark room) that captures from IR stream data.
Get the IR Stream and control the IR Emitter – Kinect for Windows SDK ...
abhijitjana.net/2013/01/11/get-the-ir-stream-and-control-the-ir-emitter-kinect-for-windows-sdk/
abhijitjana.net/2013/01/11/get-the-ir-stream-and-control-the-ir-emitter-kinect-for …


Get the IR Stream and control the IR Emitter – Kinect for Windows SDK

https://abhijitjana.net/2013/01/11/get-the-ir-stream-and-control-the-ir-emitter-kinect-for-windows-sdk/#:~:text=The%20Kinect%20sensor%20returns%2016%20bits%20perpixel%20infrared,dark%20room%29%20that%20captures%20from%20IR%20stream%20data.

 

[a a]

On Monday, 15 August 2022 at 14:51:32 UTC+2, Ed Lee wrote:

On Sunday, August 14, 2022 at 1:56:01 PM UTC-7, a a wrote:
On Sunday, 14 August 2022 at 18:24:59 UTC+2, jla...@highlandsniptechnology.com wrote:
On Sat, 13 Aug 2022 13:42:30 -0400, Phil Hobbs
pcdhSpamM...@electrooptical.net> wrote:

Ed Lee wrote:
On Friday, August 12, 2022 at 3:40:38 PM UTC-7, a a wrote:
On Friday, 12 August 2022 at 23:58:11 UTC+2, Ed Lee wrote:
On Friday, August 12, 2022 at 2:31:31 PM UTC-7, a a wrote:
On Friday, 12 August 2022 at 22:37:28 UTC+2, Ed Lee wrote:
On Friday, August 12, 2022 at 1:06:35 PM UTC-7, a a wrote:
On Friday, 12 August 2022 at 21:55:26 UTC+2, Ed Lee wrote:
On Friday, August 12, 2022 at 11:45:18 AM UTC-7, a a wrote:
On Friday, 12 August 2022 at 15:46:11 UTC+2, Ed Lee wrote:
On Friday, August 12, 2022 at 3:23:57 AM UTC-7, a a wrote:
On Friday, 12 August 2022 at 12:18:44 UTC+2, a a wrote:
what chip, clocked at what frequency, is used to measure Time of Flight of laser light at such short time intervals ?
GHz ASIC
3E-9 second per meter
c = 299 792 458 m/s > >>> c = 300 000 000 m/s

3E-9 s/m x 300 000 000 m/s = 0.9

ok
but for 1 cm resolution
we need 100 x faster clock

and for parallel analysis of point cloud 100 x 100
we need 100 x 100 faster clock

could you explain ?
There is no point in measuring all 10,000 points all at once..
so do you suggest,
what is marketed by iPhone and called Lidar in smartphone,
is a single point Laser Range Meter functionality ?

If you are correct, so why do they present 2D laser scanner functionality on images ?

Single point laser Lidar requires rotating head to work

\\so some kind od mechanics is involved
Yes, precisely, to have better relative positioning with multi-sensors. However, there is no need to measure them all within E-9 second.
-E-9 second cloc k is for 1m resolution
for 1 cm resolution
you need
E-9 * 10-2 second clock

but we still discuss a single-point operation
With 5GHz (close to current fab cap) hardware counters, we can count pulses to the same point 100 times. If we get 25 pulses of 1 meter and 75 pulses of 1.1 meter, we can guess that the distance is close to 1.025 meters.

Give me $100,000 and i can build you the chip to prove it.
-==c = 299 792 458 m/s > >>> c = 300 000 000 m/s

-===3E-9 s/m x 300 000 000 m/s = 0.9

300 000 000 m/s = 100 * 300 000 000 cm/s

= 30 000 000 000 cm/s

so you need 30 GHz clock x 2 to get 1 cm resolution for a single point

3GHz (reasonable clock) for 10cm resolution.

ok, for 1 m distant object
300 MHz x 2 clock can do the job

but if you need mobile laser range meter to scan objects on the fly, to act as 2D scanner,
1 cm counts and makes the difference

https://en.wikipedia.org/wiki/Oversampling
for higher resolution.


The usual method is a time-to-amplitude converter.

Cheers

Phil Hobbs
One way is to use a simple analog time stretcher and a
modest-frequency counter.
just watched video

kinect - sensor IR projection

https://youtu.be/MlTf0yYQjSg?t=43

let me know if the projected IR points cloud is made of static preset points
and what is a role of the pattern ?

Can theory behing Kinect be turned into large scale indoor 3D scanner,
replacing lenses in ca,meras,
moving IR projector far from VGA camera to read depth at large distance ?

Can IR pattern projected be replaced by another, higher IR power/ higher resolution pattern ?
Nobody\'s talking, so i will.

This IR stuff has nothing to do with measuring distance, unless you have to do it in the dark. If you are asking about 3D distance mapping, you have to enable it one at a time, even if you have a 100x100 laser matrix.

As someone pointed out, you can buy existing chips. But if you are integrating it into laser matrix, you might have to build your own hardware counters and routing circuits. 5GHz counters can handle at least an inch or couple cm resolution, before oversampling. You can turn off most of the circuit except for fraction of a second, in order to save power.

from
https://www.dfki.de/fileadmin/user_upload/import/8767_wasenmuller2016comparison.pdf

The Kinect v1 measures the depth with the Pattern Projection principle, where
a known infrared pattern is projected into the scene and out of its distortion
the depth is computed. The Kinect v2 contains a Time-of-Flight (ToF) camera
and determines the depth by measuring the time emitted light takes from the
camera to the object and back. Therefore, it constantly emits infrared light with
modulated waves and detects the shifted phase of the returning light [17, 18]. In
the following, we refer to both cameras (Pattern Projection and ToF) as depth
camera

not sure how ToF can be live implemented into 2D image analysis