- Видео 74
- Просмотров 223 828
LUCID Vision Labs
Канада
Добавлен 18 дек 2017
LUCID Vision Labs™, Inc. designs and manufactures innovative machine vision cameras and components that utilize the latest technologies to deliver exceptional value to customers. Our compact, high-performance GigE Vision cameras are suited for a wide range of industries and applications such as factory automation, medical, life sciences and logistics. Our expertise combines deep industry experience with a passion for product quality, technology innovation and customer service excellence. LUCID Vision Labs, Inc. was founded in January 2017 and is located in the greater Vancouver area, Canada.
Powering a Machine Vision Camera
This is quick beginners video showing the different ways to power a machine vision camera.
0:00 Intro
0:13 PoE using Network Interface Card
0:55 PoE using a Network Switch
1:39 PoE using a PoE Injector
2:05 Using External Power via the GPIO port
2:43 PoE Standards that LUCID cameras support
Industrial Machine Vision Cameras thinklucid.com
© LUCID Vision Labs
#MachineVisionCamera #IndustrialCamera #LUCIDVisionLabs #MachineVision
0:00 Intro
0:13 PoE using Network Interface Card
0:55 PoE using a Network Switch
1:39 PoE using a PoE Injector
2:05 Using External Power via the GPIO port
2:43 PoE Standards that LUCID cameras support
Industrial Machine Vision Cameras thinklucid.com
© LUCID Vision Labs
#MachineVisionCamera #IndustrialCamera #LUCIDVisionLabs #MachineVision
Просмотров: 144
Видео
Triton2 EVS: Event-Based Machine Vision Camera
Просмотров 1,9 тыс.2 месяца назад
LUCID's Triton2 EVS camera features the Sony / Prophesee 0.9MP IMX636 and 0.3MP IMX637. These cameras provides event data instead of frame-based data. For Event-Based sensors, each pixel intelligently activates itself depending on the contrast change (movement) it detects. This enables the acquisition of only essential motion information, continuously. There is no framerate anymore. Triton2 EVS...
Wafer Inspection using Triton2 SWIR - 2.5GigE camera with 5.2MP Sony IMX992 SenSWIR Sensor
Просмотров 2,3 тыс.3 месяца назад
This video showcases LUCID Vision Labs' Triton2 - 2.5 GigE SWIR camera featuring Sony's 5.2 MP Sony IMX992 SenSWIR™ sensor. Watch how we inspect different wafers and see how this second-generation SWIR sensor compares to Sony's first-genenation SenSWIR sensor. Triton2 SWIR with 5.2MP IMX992 thinklucid.com/product/triton2-swir-5-2mp-model-imx992/ Triton2 SWIR with 3.2MP IMX993 thinklucid.com/pro...
Time of Flight 3D Camera Comparison: Helios2 Wide vs Helios2 point cloud
Просмотров 1993 месяца назад
Here is a quick side by side comparison showing the difference in area that can be 3D imaged by Helios2 Wide (FoV: 108° x 78°) vs Helios2 (FoV: 69° x 51°) Helios2 Wide: thinklucid.com/product/helios2-wide-time-of-flight-tof-ip67-3d-camera/ Learn more about LUCID's 3D Cameras at thinklucid.com/helios-time-of-flight-tof-camera/
IEEE-1588 PTP (Precision Time Protocol) JupyterLab Notebook for LUCID Cameras
Просмотров 9014 месяца назад
Download PTP Notebook: thinklucid.com/jupyterlab-resource-center/ PTP and Action Commands: support.thinklucid.com/app-note-multi-camera-synchronization-using-ptp-and-scheduled-action-commands/ Intro to PTP: support.thinklucid.com/knowledgebase/precision-time-protocol-ptp/ PTPSync and Bandwidth Sharing in Multi-Camera Systems support.thinklucid.com/app-note-bandwidth-sharing-in-multi-camera-syst...
Visionary Machine's 3D Spatial Imaging System using LUCID Vision Labs Triton IP67 Industrial Cameras
Просмотров 7996 месяцев назад
The collaboration between Visionary Machines and LUCID exemplifies how innovative sensor technologies can revolutionize machine perception, providing solutions that surpass the limitations of existing modalities. Pandion™’s real-time 3D perception emerges as a game-changer in spatial awareness, setting new standards for efficiency and safety in various industries. Visionary Machines - visionary...
3D Time-of-Flight (ToF) Cameras: Helios2 3D Cameras Compared
Просмотров 2,5 тыс.6 месяцев назад
In this video we give an overview of the similarities and differences of the Helios2 Time-of-Flight (ToF) 3D cameras. The Helios2 3D cameras are perfect for automated assembly, bin picking, inspection, mobile robotics, position and location identification, and recognition and object classifications. 0:00 Helios2 Models 0:31 Model Similarities 1:41 Distance Modes 1:56 Filtering Controls and Exam...
AI Controlled Autonomous, Electric Passenger Ferry Uses Triton HDR Cameras
Просмотров 7357 месяцев назад
Read the full case study at thinklucid.com/case-studies/autonomous-passenger-ferry-uses-triton-hdr-cameras/ Zeabuz, in collaboration with the ferry operator Torghatten, successfully launched the world’s first commercial autonomous ferry in Stockholm in the summer of 2023. Zeabuz employs a comprehensive suite of sensors, including cameras, LiDAR, Radar, and AIS. These sensors provide input to ob...
SWIR Camera Coffee Bean Inspection - Tutorial: SWIR Camera, Lens, Lights
Просмотров 2,1 тыс.10 месяцев назад
SWIR Camera Coffee Bean Inspection - Tutorial: SWIR Camera, Lens, Lights
Standard Camera vs Triton HDR Camera with AltaView Tone Mapping
Просмотров 64911 месяцев назад
Standard Camera vs Triton HDR Camera with AltaView Tone Mapping
Photonics Media Webinar: RDMA for High-Speed Cameras. Optimal Image Transfer Over 10GigE
Просмотров 531Год назад
Photonics Media Webinar: RDMA for High-Speed Cameras. Optimal Image Transfer Over 10GigE
MVPro Media's Matt Williams speaks to LUCID's' Torsten Wiesinger (EMEA Region)
Просмотров 340Год назад
MVPro Media's Matt Williams speaks to LUCID's' Torsten Wiesinger (EMEA Region)
HDR Tone Mapping on the Camera: LUCID's AltaView On-Camera Engine
Просмотров 1,4 тыс.Год назад
HDR Tone Mapping on the Camera: LUCID's AltaView On-Camera Engine
Roboshin - Multi-stack pick up gripper and arm using Helios2
Просмотров 887Год назад
Roboshin - Multi-stack pick up gripper and arm using Helios2
LUCID SWIR and UV Cameras - inVISION Webinar: Spectral Imaging Presenation
Просмотров 705Год назад
LUCID SWIR and UV Cameras - inVISION Webinar: Spectral Imaging Presenation
Conventional Camera vs HDR camera (Triton HDR with Sony 5.4MP IMX490 CMOS)
Просмотров 1,5 тыс.Год назад
Conventional Camera vs HDR camera (Triton HDR with Sony 5.4MP IMX490 CMOS)
Sony SWIR Sensors & TEC Cooling: Atlas SWIR and Triton SWIR Cameras
Просмотров 1,6 тыс.Год назад
Sony SWIR Sensors & TEC Cooling: Atlas SWIR and Triton SWIR Cameras
2.5 Gigabit Ethernet (2.5GigE, 2.5GbE) Triton2 Industrial Camera
Просмотров 2 тыс.Год назад
2.5 Gigabit Ethernet (2.5GigE, 2.5GbE) Triton2 Industrial Camera
2.5GigE, Event-Based, 10GigE with RDMA, 25GigE Industrial Cameras Coming 2023
Просмотров 791Год назад
2.5GigE, Event-Based, 10GigE with RDMA, 25GigE Industrial Cameras Coming 2023
VISION 2022 Presentation: Advantages of JupyterLab for Machine Vision
Просмотров 401Год назад
VISION 2022 Presentation: Advantages of JupyterLab for Machine Vision
Lens Mounts Compared: Phoenix Machine Vision Camera
Просмотров 1,3 тыс.2 года назад
Lens Mounts Compared: Phoenix Machine Vision Camera
TensorFlow Object Detection Jupyter Notebook - JupyterLab in ArenaView Tutorial
Просмотров 3,8 тыс.2 года назад
TensorFlow Object Detection Jupyter Notebook - JupyterLab in ArenaView Tutorial
Atlas10 (10GigE Machine Vision Camera) - Sony 47MP IMX492 Rolling Shutter Sensor Overview
Просмотров 2,4 тыс.2 года назад
Atlas10 (10GigE Machine Vision Camera) - Sony 47MP IMX492 Rolling Shutter Sensor Overview
HDR Imaging for Automotive Sensing Applications - Sony IMX490 CMOS Sensor
Просмотров 3,5 тыс.2 года назад
HDR Imaging for Automotive Sensing Applications - Sony IMX490 CMOS Sensor
Jupyterlab for Machine Vision Cameras in LUCID ArenaView (with Barcode Reader Example)
Просмотров 3 тыс.2 года назад
Jupyterlab for Machine Vision Cameras in LUCID ArenaView (with Barcode Reader Example)
LUCID Triton Edge - AMD Xilinx MPSoC Embedded Vision Camera - On-Demand Webinar
Просмотров 7352 года назад
LUCID Triton Edge - AMD Xilinx MPSoC Embedded Vision Camera - On-Demand Webinar
Atlas 5GigE (5GBASE-T) Machine Vision Camera
Просмотров 12 тыс.2 года назад
Atlas 5GigE (5GBASE-T) Machine Vision Camera
TCP for 10GigE Machine Vision Cameras: Reliable Image Transfer for the Atlas10
Просмотров 9702 года назад
TCP for 10GigE Machine Vision Cameras: Reliable Image Transfer for the Atlas10
Greenholt Extension
I keep saying, configure this cameras for microscopes and grab market share. Use IMX585
😊
Very cool. Thanks for the info!
Hi, what is the name of the rods and accessories you are using for holding the light in place? 😅
Check out thorlabs!
Thanks🙌🏻
May I know the cost
Very informative video! What's the type of wafer you used? Where can I possibly get this from? Thanks!
You can find all softs or different sizes of wafers online, etched or unetched, from places such as Ebay or Alibaba.
Hi, I am using a 2 camera set up using Lucid Triton064S cameras. To implement PTPSync I have used the code provided in the JupyterLab-Resource-Center , I am facing the following two issues to set the AcquisitionStartMode and PTPSyncFrameRate # Error 1: c.device.nodemap.get_node('AcquisitionStartMode').value = "PTPSync" -------------------------------------------------------------------------- ValueError: 'AcquisitionStartMode' node does not exist in this nodemap (some suggestions): ['AcquisitionStart', 'AcquisitionMode', 'AcquisitionStop', 'AcquisitionFrameRate', 'AcquisitionFrameCount', 'AcquisitionControl', 'AcquisitionLineRate', 'AcquisitionBurstFrameCount', 'ActionUnconditionalMode', 'DecimationVerticalMode', 'AcquisitionFrameRateEnable', 'DecimationHorizontalMode', 'BinningHorizontalMode', 'ActionGroupKey', 'LineActivationVoltage', 'SequencerMode'] -------------------------------------------------------------------------- # Error 2: c.device.nodemap.get_node('PTPSyncFrameRate').value = FRAME_RATE --------------------------------------------------------------------------- ValueError: 'PTPSyncFrameRate' node does not exist in this nodemap (some suggestions): ['TransmissionFrameRate', 'AcquisitionFrameRate', 'AcquisitionFrameRateEnable', 'GevPAUSEFrameReception', 'FirmwareUpdate', 'PixelDynamicRangeMin', 'PixelDynamicRangeMax', 'GevSCPSDoNotFragment', 'BalanceRatio', 'TransferPause', 'AwbStatsFrameCount', 'TriggerLatency', 'TransferStatus', 'SerialBaudRate', 'PtpOffsetFromMaster', 'GevPAUSEFrameTransmission'] ---------------------------------------------------------------------------
Hi! Please email this information to support(at)thinklucid.com
We got predator dinosaur vision before GTA 6.
4 minutes of information. 👍🏼
Dont get it... You could do this processing with Software or?
No, because then you're back to processing image frames from a standard camera. Event-based vision doesn't send image frames, doesn't have an FPS. The data stream dynamically changes based on the amount of pixels detecting brightness changes.
What you see on the monitor is what you get without processing. You get motion information for free.
You guys are dope!🎉😮
We think you're dope too liltnt!
do you have example code real-time point cloud stream?
Example codes can be accessed by downloading our Arena SDK from our website.
Hi can you comment on Triton2 12.3 MP Model (IMX304) camera
Please make video on the integrating it with python by using sdk
These comparison videos are invaluable keep it up!
My interest is the detection of bruising in fresh fruit. The spectral band of the camera includes 900nm and the custom light at these bands will greatly matter in our fresh fruit work. This could be a possibility for fresh fruits outgoing inspection. We need to contact the supplier in the ASEAN region if any and enter into a proof of concept. Is there a demo you show using fresh fruits?
Is it possible to upload a video regarding how to calculate distance using python for Helios 2 .
Hi, the code thing at the end of the clip looks like what we are looking for. Can it be setup, that the full frame view will be displayed with the camera image by firing up the computer or open a link from the desktop? We are looking for a camera view on a flatsceen as a window replacement.
Yes, this is possible. You can set the program's window size to specific dimensions or full screen. The code at the end is a JupyterLab Notebook for our camera viewer software (ArenaView). Our full Arena SDK provides other APIs as well to create custom camera software for you application.
Thank you. could you please show an example of combining RGB camera with Helios?
Here are two 3D models with RGB+3D sketchfab.com/3d-models/assorted-objects-helios2triton-cameras-73d4ef94789c41e8a99bf40fdbb7b72b sketchfab.com/3d-models/laundry-detergent-bottles-helios2triton-750769440e8c47edb18995ca1f008feb
Do you have an example of showing the overlaying of color camera with the TOF camera? I appreciate your input
Here are two 3D models with RGB+3D sketchfab.com/3d-models/assorted-objects-helios2triton-cameras-73d4ef94789c41e8a99bf40fdbb7b72b sketchfab.com/3d-models/laundry-detergent-bottles-helios2triton-750769440e8c47edb18995ca1f008feb
I need a ToF camera below the 850nm. Why cant i find any product for my application 😔
That's because any VCSELs below 850nm will conflict with many types of indoor lighting and ambient light. 850nm is used because it is outside the visible spectrum. ToF cameras have to filter out ambient light and using a wavelength lower will end up getting filtered too.
Whats the price of the camera? I dont find the information anywhere.
Atlas SWIR 1.3MP with TEC $14,500 Atlas SWIR 0.3MP with TEC $9,950 Triton SWIR 1.3MP without TEC $10,950 Triton SWIR 0.3MP without TEC $6,950 (all in USD)
you dont have to buy such as expensive camera as 10,000 $ every camera has infrared spectrum, you need to remove inrfared filter and put infrared pass filter
@@bg7293 This is only a little bit true. If you remove the IR filter on a standard color sensor, you are looking at increased sensitivity from the 700nm to ~900nm, with sensitivity falling off to near 0% QE around 1000nm. These SWIR cameras provide excellent sensitivity far above that, up to 1650nm.
Great Video. Love your products And good to see you diving into embedded systems BTW, may u share the slides with us? thanks
Very useful. I would like to hear more about Lucid cameras.
Which assembly frame you used?
Check out Bosch Rexroth aluminum profiles
Finally.... I no longer have to drink plastic coffee....
Fun fact! Did you know that the global coffee beans economy was valued at around $31.93 billion in 2022 and is expected to grow at a CAGR of 6.8% from 2023 to 2029, reaching a value of $50.61 billion? Detecting foreign substances in coffee beans is a serious matter! There are lots of opportunities for plastic bits (and other bad guys) to fall into bean harvests and SWIR cameras are used to detect and remove them!
Can this detect a jelly belly in a bowl of raisins? I've been trying to figure this one out for years and finally it looks like I might just be able to do that.
There is a good chance it could! SWIR cameras are really good at detecting the water content in small objects. If the water content is different between the jelly bellys and the raisins there is a good chance that one would appear darker (or lighter) that the other, irrespective of jelly belly color.
I’m never ever drinking coffee again
Very informative.
Costs?
USD $6,950.00 (Triton 0.3MP SWIR) up to USD $14,500.00 (Atlas 1.3MP SWIR)
Does 10GigE network card compatible with 5GigE cameras? What speed will be maintained&?
It depends, some 10GigE cards may not support NBASE-T speeds (5 & 2.5GigE). If it does support NBASE-T speeds, the 5GigE camera will run at 5GigE speeds (~600MB/s). Please check your 10GigE card's datasheet or website for more info.
good job with 490. good dynamic range gain between a two systems, different white balance makes it harder to focus on actual improvement. it would be interesting if you compared with same generation sensor from onsemi, such as ar0323,
Hi, nice camera collections. please, do the camera support or provide raw uncompressed pixel by pixel data? I have need of camera but I need to access raw pixel data of the camera. Also which of these best fit for range from mid IR (1000nm) to near UV (300nm).
You have mantioned on 1:50 corrupted frams/lines and so on. i have a very strong PC with intel i9 12k, 64 ram 1 gige Cognex 4k camera. and on my high speed application i have corrupted images sometimes. On next image there is a part of old image. i run only with one camera, tryed 4 different PCs, 2different cameras. Using basler gigevision driver. In pylon software i can wee the using Windows Performance filter. Jumbo packets 9k.. tryed everything but cant locate what couses the errors. No matter image size, no matter FPS (10 or 150) its still happening. Help pls
Sorry to hear about your issues. Are you sure the Basler gige driver is the proper driver for your Cognex cameras? If you feel you've done everything correctly the only thing you didn't mention you tested is your cables. Are you using the same cables when you tried different computers? Poor quality cables can absolutely cause frame corruption and is often overlooked as a source of issues.
@@LUCIDVisionLabs Hi, thanks for suggestions. Yes, ive tested it with different cables. I ve found the probable issue, but can not solve it. I am using Halcon library with C# code. WHen i run my application (winforms), it start doing that corruption stuff when i try to move my winforms window or somehow interact with PC or windows (probable cpu thread starvation?). When i run the same code but only in Halcon in Hdevelop, it does not corrupt frames all is OK.. But PC has top hardware and i still can not figure it out what cousing those corruptions.. If cpu got freezes they why frames are corrupeted and lines image lines are populated with image line from different image,,Strange. Thanks for you assistance
Tf1?
can this camera work with raspberry pi?
Yes! We have tried a total of 4 cameras on the Raspberry Pi, (1 via the Ethernet connection + 3 more cameras with an Ethernet adapter on a USB2 port). The 3 cameras using the Ethernet-to-USB2 adapter have limited bandwidth however.
@@LUCIDVisionLabs Is this running Ubuntu for ARM version of ArenaView? Do you have anymore resources for running this on a RPi?
@@zaidparkar8810 Unfortunately we don't have an ArenaView GUI yet for Ubuntu. We just have our SDK (APIs) for ARM Ubuntu 18.04/20.04/22.04, 64-bit
0:13 The distortion looks strange. Assume he is bouncing the ball at 0.5 m amplitude, 2 bounces per second. At the apex, the ball is not moving faster than about 2 m/s. During 62.5 us exposure, the ball travels 0.125 mm, which is practically perfectly still given the pixel size at this target distance. Why the distortion, then? Image Accumulation is off, so I'm assuming that each single frame produces a distinct point cloud.
This is because 1 point cloud frame is derived from 4 phases (4 micro frames). If the object is moving too fast and changing positions for each micro-frame it will become distorted. More info here: thinklucid.com/tech-briefs/sony-depthsense-how-it-works/
@@LUCIDVisionLabs That page states: "It is only necessary to have the 0° and 90° micro-frames to calculate depth.". Do you provide 2 micro frames mode?
@@PaulJurczak Unfortunately we don't. Our Normal mode uses all 4 micro-frames (2 for distance calculation, and the other 2 to refine that calculation). In High-Speed mode we use only 1 micro-frame (provides much less distortion, higher FPS, but at the expense of accuracy and distance range)
@@LUCIDVisionLabs What is the total time of acquiring 4 micro-frames at 62.5 us exposure?
The total time for the 4 micro-frames, ignoring the idle period (grey part) is about 12.4ms, or 12400us with 62us exposure. The readout + reset time (blue and green parts) is just under 3ms for each of the 4 phases. (This comment is referring to the 1 frame diagram in the Sony DepthSense article.)
Great video as always! Can you estimate how much power the tone mapping required?
Hi Gal, the TDR054S with AltaView has a power consumption is 3.8W via PoE and 3.2W via GPIO. So it takes up a tad more from our Triton camera average which is around 3.5W via PoE, 3.1W via GPIO)
Thanks for this video I feel like I'm an expert now
Lets be experts together! 🤝
Can I live stream with OBS and get a similar result to the guy welding?
It is possible. My understanding is that OBS will stream whatever is on your monitor, full screen or windowed. So you could build a camera viewer program that's doing HDR processing using our Arena SDK APIs, then launch that program and then use OBS to stream that program.
FYI we don't support DirectShow.
awsome explanation
What about SWIR skin penetration?
SWIR light can penetrate into deeper levels of skin tissue.
But its not going through the apples at longer wavelengths!
Shortwave infrared light not go completely though the apples.
@@LUCIDVisionLabs But many studies have reported SWIR light going through bone (skull) and reaching the brain for therapy applications. So it should penetrate apples also.
@@pulkitsharmapremiumvideos9252 We don't sell SWIR lighting so I can't comment on what SWIR lights which were used in those tests. LUCID's Atlas and Triton SWIR cameras do not emit SWIR light and are only sensitive to it. So it is important to do your own testing with different lighting and optics with our camera to maximize the performance of your specific application.
Literally everything EXCEPT how a indirect ToF image sensor works..... Direct single point ToF and static point cloud depth sensing make sense to me, but this real time ToF across an entire image frame without capturing frames in the picosecond range, Im completely lost. I though LUCID could help....but I guess I was wrong.... [walks away, sad, in the rain]
There is a delay between the light sending and returning and the camera calculating the point clouds. The delay is around 10 - 12ms.
Very detailed information, thanks !
The camera doesn't do tone mapping onboard I assume?!
Correct, at the moment the camera can send 24-bit RAW image data to the PC. Then you can apply your tone mapping algorithm running on the host PC.
@@LUCIDVisionLabs It's very cool. We're doing 3D reconstruction scanning along orchards/vineyards with pretty high optical flow, so I think the rolling shutter is a problem. Do you know how fast the readout is top-to-bottom? (Does that have a proper name)?
''' THANKS'' SUPERRRRRRR....
''' THANKS..........!
Too difficult to get a quote, went elsewhere