[Music] welcome to evolution in controls I'm your host Tim Wilson thank you for joining us in the fast-paced world of Industrial Automation it's become commonplace to talk about packages or products moving faster and faster in some respects it gets easier and easier to make them go faster but what doesn't get as much attention is that as it's easier and easier to make them go faster it gets harder and harder to gather accurate information about them as they travel speed and efficiency aren't just about the mechanical functioning of machines it is also about the accurate and consistent recognition collection and recording of data about the products and packages it's ironic that the visual collection of information in a product delivery system is often the Unseen element in the past the barrier to faster systems was the mechanical mechanism that move the product today the challenges data management how do we collect organize and manage the data stream it's no good moving products faster if we can't capture and handle the information about the product enter the scan tunnel a solution designed for the collection of a significant amount of visual data that outperforms even the fastest of the Delivery Systems and it's not just about the collection of a significant amount of visual data the speed of the collection must be matched by the speed of the process processing of that data this isn't your average scanning system but how does this technology transform the way we approach data management in industrial settings here to talk about the scan tunnel as not just a tool but as a GameChanger in the realm of automation is Mr Mark Baron Mark is the founder of simple vision technology and he's going to help us answer questions about the capabilities of the scan tunnel mark thank you for joining us thanks for having me why don't we start tell us tell us a little bit about yourself and and then maybe simple vision technology so I have over 27 years experience in the automation manufacturing industry I have over nine years of experience in the uh Machine Vision industry and just over three and a half years ago we started simple vision technology as a software company okay so what does simple vision technology do simple vision technology is a single software user interface that is Hardware agnostic that is used to compute and decipher images that are captured by industrial cameras so do you make cameras also no we just make the software we focus on making the user interface for the software very comfortable and easy for them to use doesn't matter what cameras you you use we we do not care about the camera that is used we care about processing the images as best as we possibly can in the p PC we don't uh require any specific Hardware other than a PC that would run the software software yeah okay well we're here to talk about scan tunnel which is a a test bed for vision and vision technology and and apparently software as well um what are we going to find when we go out there's a picture of it here so what are we going to find I see a mechanical structure here it looks like it's got a pallet conveyor and and maybe two elevators yeah so this this system here is made up of two different elevators and two different pallet conveyors that allow them to circulate pallets in a around continuous motion and what we can do is we can actually acquire images and use that demo as a pseudo scan tunnel we can add cameras to it or we can take away we can add shrouding and change lighting but we can really use this as a a test so the the scan talk I mean you're separating mechanical from Vision I guess yeah so it's really not about the mechanical structure this is really about the vision technology that we're employing here but if we're talking about a scan tunnel it's really two components there's the mechanical component and then there's the vision component what I call Vision component being the uh acquiring of an image setting up the lighting processing the image and then disseminating that information out to some supervisory control that would allow a parcel or whatever it is coming through a scan tunnel to be uh manipulated what's the mechanical piece the mechanical piece is creating the structure that covers the maybe a conveyor in this case that creates a dedicated lighted space envir environment yes it creates a dedicated environment that allows us to have repeatable images so that we can really decipher and decode images as quickly and as best as possible at the heart of this test bed though is the vision technology yes yes so we we utilize industrial cameras industrial Computing industrial structures to create the scan tunnel and then your software is on this technology so we we talked a little bit about about cameras you said you were Hardware agnostic I think what you said um there are cameras though that already do image process it's on the edge processing they're they're they're capturing the image they're processing it right there on the camera are they doing the same thing that your system is doing so what they do with an edge device is that device has a computer or a processor internally it has a program that is used to decipher an image and it has an image acquisition device or a camera built in camera so that camera acquires the image and creates a output to the super supervisory system of past fail good bad or whatever the results might be we at simple vision technology created a software interface that will compute and do everything that the edge devices do but we do it in a central location so what this hap what what this helps with is the deployment of multiple cameras or the management of data and the ability to configure all of the cameras and the the features and fun functions in a central location you're talking about the cost of scale up or the ease of scale up correct yes because you have an edge device that is a fixed price and so each one is a linear fixed price increase where when you provide a PC with software and you add a couple of cameras then as you add these let's call them dumb cameras cameras that only are used for acquisition the cost of them is so effective that it really dramatically drops the price on our test bed here mhm smart cameras or dumb cameras dumb cameras you currently have four dumb cameras here but could easily add another 10 cameras if needed and not by changing any of the processor or by changing anything else but really just adding dumb cameras tell you what let's let's break here let's go out to the lab and have a look at this and we'll we'll get back together and talk about it that'd be great okay we'll do that thanks all right Mark we're out here in the shop and we're next to the uh test bed for vision and vision processing in your technology simple Vision Technologies we we've got an extruded aluminum structure here and on on it's a conveyor system a belt and pallet style conveyor there's an elevator on each end of it and we've got some pallets on the conveyor with some paper that's got some QR codes on it we're standing next to one of the conveyors and above the conveyor and the elevator we've got two lights a red light and a white light and it looks like we've got two cameras they look identical but one's got some kind of lens on it or something I you'll be able to tell us about and then we've got a a TV screen over here to my left that is showing me what the cameras are seeing correct yeah this is the simple vision technology software interface all right so what are we what are we going to be seeing here when this when this runs what's going to happen so so in this demonstration we have a pallet that is going up and down okay and we have it's the elevator it's the elevator of this of the demo and it's identical on both ends as you mentioned right there are two cameras right one camera has a fixed focal length lens which means we focus it on a certain plane and certain distance a certain distance away from the camera and that's where it will only be in Focus so as it's because it's fixed Focus that means when the pallet is in one position it's going to see it in Focus but as that pallet moves up or down it's not going to be in Focus anymore correct and so with that camera we're going to take a picture at the raised position and it will be autofocus and then when it fully lowered the elevator will take another image and you'll see that it is in Focus so it's not a motion it's not a a continuous video camera no it's sending still shots to your technology correct exactly and some are in focus and some are not so your technology is making the decision on when it's in focus and then it's capturing that still it's capturing the image transferring it to our system and then it's deciphering if it can find a code in the image or not the content of the content of that yeah all right so that's the fixed focal length camera correct what's the other camera the other camera is showing the technology of a liquid lens right so is the camera are the cameras identical the camer except for the lens the cameras are identical except for the lens correct and that is a liquid lens this liquid lens is a technology that allows you to continuously remain in Focus no matter what plane you've selected so the lens is changing its focal length exactly H how is it doing that you said liquid filled is that there's a technology there that makes the focal length of the lens change there is there's a technology that it will require a voltage to change the focal length of the how how does it know what focal length to have is it just continuously moving up or down until it finds the the focus no so what we've done here in this for this demo is we've provided a a laser sensor is that the Red Dot that's the Red Dot that you see here that you see on the screen on the screen there and there and what we're doing is we're measuring the height of the pallet so that we can provide a voltage that allows us to stay in focus no matter what plane we're on so it's a closed loop system measuring the distance and as the distance changed as measured by the laser yes it adjusts the focal length on the camera exactly and it has nothing to do with the software it is really just the closed it's just the lens yep it's the lens and the and the measuring system all right and both of these cameras are feeding into a controller in the back of the conveyor here correct correct yeah and your software is running on that Hardware device exactly yes we we are Hardware agnostics so we can utilize multiple industrial strength cameras that are very costeffective to install when we were in the conference we talked earlier about Edge processing or where the where the processing done is is any of that happening on these cameras are they smart devices no so these devices are actually image acquisition only and they transfer the data or the image directly back to the simple vision technology system so this while this may be what it's seeing it's not processing any of the the cameras aren't processing this data correct the process data is done directly in the controller where the SVT software resides all right when we start the conveyor up what are we going to see in the pictures on the TV yeah okay so the top right picture there yeah that will be the fixed Focus lens and we'll see that it's out of focus when it's raised and we'll see that it's in focus when it's lowered all right and we'll see two different images so will I actually see this go out of so this is a still image right correct and and once the conveyor starts moving it's going to go out of focus correct yes uh when it's not in the correct focal plane the image will be out of focus and then and then once it gets in Focus you're saying that then it it captures that and it Remains the still that's why it's a still Focus exactly the bottom is is continuous yes we have that in what we call a live video mode and so we're able to run this camera in a live mode you'll see as the pet enters and exits that the the video will stay live but what's also going to happen is the elevator will go up and down always stay in Focus as the elevator goes up and down yeah exactly all right I think we're ready let's go ahead and start the conveyor now so it's going down out of focus and we're seeing staying in Focus the whole time oh it snapped into Focus there it did because the fixed focus is at the bottom now why is it in why I don't know if we can see that but it's in green why is that in green so that depicts that it found a data Matrix code in this demo we're showing that we can find a data Matrix but we don't necessarily want to find the QR code in this image so the green is depicting that this is the code that I is what it was looking for this is what I was looking for and this is your software on that Hardware device that's not the camera doing this correct it's all the software that is deciphering the image to provide a result now this is a test bed so all we're really doing is looking at at pieces of paper but if this was production if there was Data inside that green box would that data be sent then someplace absolutely to whatever industrial controller uh that is chosen for the project could it do something as as as simple as checking orientation of a box or POS of a package on the con on the pallet yes so the simple vision technology software has the ability to do barcode reading rule-based vision and artificial intelligence from this Central Lo located uh interface okay and so therefore yes if we wanted to read just codes we it's possible if we want to do rulebase Vision as well or any kind of AI inference comparison it's completely possible now I'm going to say these conveyors are moving slow if this is not a fast F system if I had boxes flying down some kind of conveyor are you going to be able to keep up absolutely yes so what we do is we allow the user interface for the user to create um these scenarios that allow them to get the optimal image so that they can read the codes that they're looking for but if if I've got if I've got a conveyor that's going feet per second how fast how fast can these cameras capture the image communicate it and then process it so depending upon the camera that's chosen there are multiple frames per second that are acquired so we have the time to capture images transfer them to the industrial uh PC that then deciphers the codes so in Practical terms even the fastest conveyor is going to be not fast enough to outdo the camera correct and if if if required we can offset some of the triggers so as a package would maybe come into a scan tunnel we could look at a sensor that would allow us some delay here comes a package Get Ready Get Ready exactly all right yeah mark this is fascinating thank you so much for taking the time to come let's go back to the conference room and and and wrap up all right great thanks thank you appreciate it Mark that was fascinating thank you so much for for walking us through that um I'm still struck by a couple of things number one we're separating the capture of the image from the image processing or the data processing in my introduction I talked about data management you're capturing that information and you're processing it to me that there's the cost of the scale up and and what you do but there's also the concept of speed can you be fast enough uh our conveyor we're moving slow because we're showing the vision technology but if I wanted to shoot a baseball through the vision tunnel I know it's not set up but if I had that could you capture a number that was on that softball absolutely with the user interface giving the customers the ability to set up the system it's possible to set up a pre-trigger so I could really look at the velocity of what the ball is moving and then I could tell you how long it takes to get from the switch to the proper location in front of the camera and I can acquire the image with a global shutter which allows me to do a very still um crisp image in that is in motion and this would allow us to um provide a a very good image based upon whatever speed that we're sending so there's no there's no problem in actually capturing the image um processing the image because we're on a central processing unit is is much faster than it would be on an edge device and if you wanted to replicate this or if you wanted to increase this or add another line do we just have to add another SVT software device or is there an upper limit on the cameras that you can connect to this so each license allows you up to 25 cameras 25 cameras exactly and so therefore in this example we could potentially add another 21 cameras to this system with no need for another license even or no need for another Processing Unit so scalability is very cost effective what's the future look like where where is the software going where is the hardware going in five years from now are we going to be doing this the same way is the is artificial intelligence going to become more involved in this absolutely artificial intelligence is going to become more and more involved but I think where this scales at a very interesting level for customers is that an edge device when deployed mhm is a fixed processor with a fixed camera resolution as technology changes rapidly in this market processors are faster image acquisition is faster megapixel resolution is much higher and higher but that's more data that's more data but the software is the key to transferring that data and so as the cameras advance so will the PC and as long as the software is capable of transferring that data then we're good as as we talk about the scan tunnel this test bed has been a great vehicle to show this but there's more questions that I have about your technology I I think I'd like to have you come back and we just spend a whole episode talking about simple technology would you be open for that that' be great I would love to do that yeah explain a lot more about what we do that' be fantastic you know I I really appreciate your your work and your help here in walking through the scan tunnel and and how the images work and the data management piece of it thank you so much I appreciate your time thank you thanks for having me no problem to learn more about simple vision technology visit simpl Vision technology.com to learn more about evolution motion Solutions visit Evolution motion.com don't forget to subscribe to evolution in controls on whatever platform you use for podcasts or on YouTube for a video version so you can be updated when we release new episodes thank you again for joining us today I'm Tim Wilson and remember keep moving [Music]
2025-01-23 08:18