Episode 11 – Jeremy Bergh from IDS: Audio automatically transcribed by Sonix

Episode 11 – Jeremy Bergh from IDS: this mp3 audio file was automatically transcribed by Sonix with the best speech-to-text algorithms. This transcript may contain errors.

Jimmy Carroll:
Hi, everybody. I'm Jimmy Carroll. Tech B2B marketing. I'm here with Jeremy Berg of IDS Imaging Development Systems, and we're at the Autonomous Mobile Robots and Logistics Show, which is called Located with the Vision Show. Jeremy, thanks so much for taking some time. I'd like to talk to you a little bit about what IDS is excited about at the show. I know, for example, that you have the IDS NXT could you tell could you tell us a little bit more about that, please?

Jeremy Bergh:
Sure, so the NXT is I mean, it's basically our foray into the smart camera. It's giving you the ability to really solve so many different applications. What we're excited about is just seeing what customers bring to us in terms of problems and seeing if it's possible to solve it. So you can app your camera, you can create your own apps, vision apps and put them right on the camera. It does all the processing at the edge for you, or you can label some images. We have a software, the NXT Ocean that'll basically create the neural networks for you. You can upload, I don't know, up to 50 different neural networks right onboard the camera. We make it super easy for people to maybe don't have a ton of experience writing their own, their own networks, to get involved in AI and bring it to their manufacturing line or whatever they want to do. So that's what we're excited about.

Jimmy Carroll:
So A.I. deep learning, very interesting concept. For years now it's been kind of hyped and, you know, I think it was maybe the Stuttgart show 16 ish where where a lot of people are coming out with deep learning. And it was being touted as this magic tool that will solve everything. Now, years later, people are finding finally finding ways to to apply deep learning in the real world. What are some ways that you've seen the NXT used from your customers that you know help solve them problems, their problems?

Jeremy Bergh:
I've seen some interesting ways that we weren't really thinking of before, like anywhere where maybe something is is boring. Like a customer wouldn't want to sit there and maybe look for a light to turn on, right? So we have even simple things like that, or we have fruit inspection, you know, agricultural applications where they're even looking at either whether it's a good or a bad object or a piece of fruit, maybe what stage the plant is in the process, maybe helping to estimate yield of, say, wine production, things like that. And then, yeah, I think we've seen a couple things in quality inspection and the automotive industry as well.

Jimmy Carroll:
Yeah, Yeah. So that's interesting. So are you finding that with your customers and a lot of what you hear now from a lot of people is, is deep learning will not replace machine vision. Deep learning will be a tool in the tool belt of machine vision and they can work together to to solve new problems. Is that what you're kind of seeing with your customer base too?

Jeremy Bergh:
Yeah, I am. And you know what? We're really it still seems like nobody knows where the limits are yet. You know, it still seems like every time I talk to, you know, even a competitor that has out some products that are somewhat similar, nobody knows where the limits are and where are the applications are that they should be focused on yet?

Jimmy Carroll:
Well, in many ways that's an exciting thing. It's it should be it should be fascinating to watch the the evolution of deep learning over the next several years. I have a question on the on the ocean software platform. So one one sort of trend or theme and I guess it's an ongoing one, right? Is things becoming easier to use? No. Code, low code, things like that. Is the ocean platform something that kind of targets, you know, ease of use, let's say?

Jeremy Bergh:
Definitely. Yeah. It's you know, customers can say they want to look for, I don't know, different colors, different kinds of shirts. Right. So they have these shirts going across the conveyor belt. They put a couple of pictures of of this shirt versus my shirt, label them. And then you can basically have that neural network find and classify the shirt. And then if you want to, you can use Google Berkeley. We have our platform, and that's basically creating the logic for you in the camera. So once it finds us, you know, your shirt with a 95% probability, that's a beautiful shirt. It will switch. You can have it switch to a different neural network that says, okay, is it a good is it in good shape or is there any flaws? Right. So and that's all done super easily with people who don't even need coding experience. So.

Jimmy Carroll:
Okay, well, I appreciate that. I'd like to maybe transition a bit over to cameras. Of course, IDS is primarily known as a camera manufacturer. Yes. Walking over to your booth today, I saw some interesting things in three D So are there are some are there some developments that we should talk about from the Ensign cell line?

Jeremy Bergh:
So there are some developments. I'm not sure how much I could talk about yet. I know there's some things coming out soon that we'll be able to share with you in terms of the incentive. We keep coming out with better and better firmware. And and, you know, on the software side, it keeps getting better and better. So there's constant improvements there, you know, but. But in terms of new hardware, you'll have to wait just a little bit longer.

Jimmy Carroll:
Okay, let's tease it. I like it. In terms of the 2D area array, traditional family, the USB and gaming cameras that you have. What's new there?

Jeremy Bergh:
So we have the Warp ten, which is our ten gig platform. We have some sensors coming out there in terms of something that we had in the past that's kind of went away for a bit, but now we brought it back is something that maybe you wouldn't seem to see in Industrial Machine vision too often. It's our X camera, it's a 13 megapixel, has an autofocus lens already built in. It's pretty simple, but we found that it solved some problems and that one we have a lot of ready to go and that one's been doing pretty well for us as well.

Jimmy Carroll:
What what type of application does that target? Is it a webcam or is it something.

Jeremy Bergh:
You know, I mean, it's similar to it's like bringing a webcam into industrial. Long term availability. It's you know, it doesn't have the trigger or flash. That's the only thing that maybe takes it away from being a true industrial camera. But we've seen it in maybe kiosk applications. Right. You know, different monitoring applications where you don't specifically need the trigger or flash.

Jimmy Carroll:
Yeah.

Jeremy Bergh:
But fairly low cost. It's got the autofocus built in. So it's it's just easy to use.

Jimmy Carroll:
So you mentioned availability, right? Yeah. Supply chain issues are affecting everybody. How have you guys dealt with that?

Jeremy Bergh:
Well, we did do a redesign of some of our components, you know, basically kicking out some of the problem components that we couldn't we couldn't get regularly. And our engineering team did do a basically they they looked at and verified about 150 different alternate components besides FPGAs and Faes and those kind of things that we could use if supply chain dwindles on certain parts of our camera. So it really made us made our supply chain more robust in terms of the ability to be more flexible.

Jimmy Carroll:
Yeah, certainly seems like a prudent move. Jeremy, what have I what have I missed? What else are you guys excited about, whether it's at this show or in general?

Jeremy Bergh:
You know, we're excited to see people again. So it's great to great to come out, of course, be able to travel, get together as a team, you know, because we've been separated for a while. And in terms of, you know, we just we love meeting people, hearing about what they're trying to solve and seeing if we can be a part of it.

Jimmy Carroll:
Of course. Jeremy, thank you so much, everybody. Jimmy Carroll, Tech B2B Marketing Again, Manufacturing Matters Podcast. Comment like Share. Reach out to us at Tech YouTube.com or LinkedIn and give us some topic ideas, some feedback, whatever. Would appreciate it. Thank you all very much.

Jeremy Bergh:
Thanks, Jimmy.

Sonix is the world’s most advanced automated transcription, translation, and subtitling platform. Fast, accurate, and affordable.

Automatically convert your mp3 files to text (txt file), Microsoft Word (docx file), and SubRip Subtitle (srt file) in minutes.

Sonix has many features that you’d love including upload many different filetypes, secure transcription and file storage, enterprise-grade admin tools, transcribe multiple languages, and easily transcribe your Zoom meetings. Try Sonix for free today.