OK, just finished the archived audio. Thanks Eric and Ann.
I've been interested in some hobby type vision projects because they are inexpensive and being a hardware guy at heart, building the circuit on a breadboard is natural.
Low resolution vision systems can be put together with off-the-shelf MCU development boards, a few chips and some software.
time-of-flight sensors is something we use here all the time in Guided Wave RADAR tank level gauging. RADAR depends on precise time of flight measurements.
TOF = Time of Flight. It measure the time it takes a pulse of light to emenate from a source, bounce off an object, and be seen by a camera. Very simular to RADAR.
BDTI Senior Engineers Eric Gregori and Shehrzad Qureshi deliver a techology trends presentation on image sensors for embedded vision applications, at the late-March 2012 Embedded Vision Alliance Summit. Eric and Shehrzad discuss trends regarding conventional 2-D image sensors, as well as those supporting "3-D" (depth discernment) and "4-D" (i.e. selective focus, otherwise known as light field or plenoptic) functions.
@Eric: I'm sure you're right, but there are a wealth of video lectures there that might take some time to sift through. Can you give a one- to two-sentence description of what TOF sensors are in this chat?
Thank you Eric! Table on slide 22 shows the "Time of Flight" type of 3D sensor that was not covered in your lecture... Are you planning to talk about it in the future lectures?
@Eric: Great lecture, and excellent slides. One question I had (and I realize I may need to resubmit this tomorrow) is that "Time of Flight" 3D sensors show up on the Summary slide (slide 22) without previous reference, introduction, or comment in the audio. Would you care to describe TOF sensors and distinguish them from Structured Light and Stereo 3D sensors which you did introduce earlier?
Again, very sorry for the technical problem today with the audio.
You should be able to access the recorded audio now. If not, please let us know.
I recommend that everyone follow today's lecture from the recording at your convenience, and bring your questions from today's lecture to the Q&A session following tomorrow's lecture. Alternatively, you can email us your questions at info@embedded-vision.com. Also, feel free to post questions on tomorrow's classroom page prior to the start of the lecture.
Eric, etc, you may want to cover -- ah, waited long enough, you got to it -- the bit about normalizing pixels. I suspect you'll mention later about cancelling out known-defective-pixels...
@wwbrown, if you click the "Play" button on the red block at the top, it'll start playing the archived lecture. You may need to click F5 to refresh and re-sync your browser window.
Listening to it right now... Just finished going through the slides (MUCH prefer the PDF instead of powerpoint; good move). Was amused -- you remember yesterday I asked about LeapMotion? They are doing a 3D sensor play with jawdropping precision (repeatable position-determination of 0.01 millimeters, ten micrometers). THEY aren't saying a whole lot about how they do it, but the only bit I'm missing at this point is whether they are using more than one CCD sensor (some element of triangulation, possibly). Infrared light, lenseless, shadowmask, calibrate with an empty field-of-view speckle pattern which is constantly applied to NEW sensor frames; the intellectual property is the algorithms which extract position and distance data from the sensor stream.
Again, very sorry for the technical problem today with the audio.
You should be able to access the recorded audio now. If not, please let us know.
I recommend that everyone follow today's lecture from the recording at your convenience, and bring your questions from today's lecture to the Q&A session following tomorrow's lecture. Alternatively, you can email us your questions at info@embedded-vision.com.
Back in the mid 80's, I implemented an embedded vision system to recognize and identify vehicles. The application was a Toll Collection system in South Korea. The drivers there recycle everything and stuff cardboard, etc between the cab and trailer or wherever they can. We took pictures of all the vehicles sold in Korea and used fourier transforms to calculate probalities that a vehicle in the lane was of a certain toll class.
Come on, guys, this is embarrassing -- I talked you up to a forum full of people interested in sensor technology and associated tools, and I'm getting razzed on a back-channel here...
You can check out any time you like, but you can never leave.
I saw a multi-Kinect system that put each Kinect on a little motion platform. Since the camera and image projector are a single unit, each one sees its own light fine, but all other light from the other Kinects is an ignorable blur. Neat solution.
@RichQ, yes, there are, but not every implementation uses them.
Even better is when you take a look, google "lenseless imaging shadowmask", you get some serious "structured light" opportunities -- and if you add in "speckle" you get even BETTER stuff. Turns out, apparently, that if you capture a shadowmasked CCD chip speckle image with nothing in the field of view, then ADD something, the CHANGES in the speckle field can in fact be processed to extract a TON of useful information.
MATERIAL TO STUDY: ANY TEXT BOOK ON SIGNAL (DIGITAL?) SIGNAL PROCESSING. WIKIPEDIA HAS QUITE A BIT OF MATERIAL, BUT A GOOD TEXT BOOK WILL HELP TO SET A SMOOTH LOGIC SEQUENCE.
@ Eric, the presentations looks good, I'll log in and listen to the archive when it is available. Thanks for setting up this course. Maybe tomorrow will work better. see you then
MAX: I wen to their site yeaterday and downoaded a cheatsheet of processor comparisons. I think the high end ARM would work.. Defintely some of the Highend TI chips
Have you dug into the OpenCV website? They have a LOT of useful info, including stuff about what platforms they are currently working with. They also have some pretty useful tool kits (I downloaded 150M zipped tool kit which allows me to monkey around with images on my PC).
@Ann: is the actual lecture being recorded now? If so, can we be notified by email when the archive is available? If not, can we be notified as to when the lecture is rescheduled?
Have you designed with image sensors or cameras in the past? If so, what was your biggest challenge?
My biggest challenge was selecting the lens ... mostly due to my lack of knowledge in optics ...
Next biggest challenge was getting a clean image ... on several occasions, with different image sensor manufacturers, there always seemed to be an obscure setting that got the image cleaned up, but was not intuitive to find in the datasheets
Can I integrate OpenCV into an Android device? I'm not familiar at all with CV and looking for starting points and what hardware to use. I'm building a car infotainment system for myself, and it needs video processing feeds.
@ANN: WHAT IS THE WORD. DO YOU THINK WE CAN CONSIDER THIS A COURSE TIME? IT IS ALREADY 2:16 pm AND MOST OF US ARE IN THE LIMBO. WHY DON'T WE RESCHEDULE THIS?
I work for Blog Talk audio and I'm here to assimilate you! Seriously I work for a large construction equipment design firm and I was hoping to see if vision systems could be applied to autonomous machine navigation
The size of the sensor changes the processing requirements.
Need to be cautious distinguishing between CCD and CMOS -- one of them gives a "full frame at once", the other one CAN give a "rolling" capture (which makes variations in lighting stand out like a sore thumb).
HAS anyone else heard about the Lytro?? It was a blip there for awhile, because with a small script download, you could take a picture generated with a Lytro device and click on any point IN the image to adjust the focus to put THAT point IN focus. Pretty cool.
Have you designed with image sensors or cameras in the past? If so, what was your biggest challenge? No Have you heard about structured light or time-of-flight sensors? Have you evaluated or used them? No
Have you designed with image sensors or cameras in the past? If so, what was your biggest challenge? No Have you heard about structured light or time-of-flight sensors? Have you evaluated or used them? No
If you're having audio issues, please note that some companies block live audio streams. If you don't hear any audio, try refreshing your browser. The show will be archived and available on this page.
Hello everyone. Based on my experimentation yesterday, Mac users should use the Google Chrome browser. The Flash Player add-on for Safari and Firefox (even the latest versions of all) seemed to not work correctly, at least for me.
Windows Firefox (and Safari) users may also be affected. IE users are presumably ok
Interesting. Eric, and not Jeff. Ahh, but I see you ARE here, Jeff. Double-teaming us, eh?
A note to "higher up" -- it appears it would be seriously worthwhile having the audio window popup with some audible "yes, you will be able to hear the official feed when it starts in ..." with a countdown, maybe??
Hey, everyone! (Did I see correctly there may be T-shirts for twenty-five of us here today? Cool!)
The streaming audio player will appear on this web page when the show starts at 2pm eastern today. Note however that some companies block live audio streams. If when the show starts you don't hear any audio, try refreshing your browser.
The streaming audio player will appear on this web page when the show starts at 2pm eastern today. Note however that some companies block live audio streams. If when the show starts you don't hear any audio, try refreshing your browser.
By the way, I don't know who else is happy about getting slides in PDF rather than PowerPoint, but I really appreciate the change. Thanks Jeff, Eric, Ann, Digi-Key, and whoever else helped make this happen.
The government wants to study your brain, and DARPA wants to use similar information to give robots true autonomy beyond any artificial intelligence developed to date. Sound like science fiction? It's not.
From Dell / Intel® New Paradigms in Design Work Scott Hamilton, vertical market strategist for Dell Precision workstations, 5/2/2013 3
Early in my career, I worked as a draftsman and remember the days of drawing on vellum with numbered pencils and Mylar with plastic lead. This was a fun experience in the sense that I ...
I've been using workstations for more than 10 years and love finding ways to get more performance from my system. With demanding professional applications that require more power each ...
A lasting memory from my first job as an engineer in an auto assembly plant is standing on hard concrete at six in the morning, vending-machine coffee clutched in hand, listening to ...
A quick look into the merger of two powerhouse 3D printing OEMs and the new leader in rapid prototyping solutions, Stratasys. The industrial revolution is now led by 3D printing and engineers are given the opportunity to fully maximize their design capabilities, reduce their time-to-market and functionally test prototypes cheaper, faster and easier. Bruce Bradshaw, Director of Marketing in North America, will explore the large product offering and variety of materials that will help CAD designers articulate their product design with actual, physical prototypes. This broadcast will dive deep into technical information including application specific stories from real world customers and their experiences with 3D printing. 3D Printing is
To save this item to your list of favorite Design News content so you can find it later in your Profile page, click the "Save It" button next to the item.
If you found this interesting or useful, please use the links to the services below to share it with other readers. You will need a free account with each service to share an item via that service.