Close Ad

The Spoon

Daily news and analysis about the food tech revolution

machine vision

How California’s New Food Waste Law Could Catalyze Interest & Investment in Tech-Powered Food Recovery Platforms

Starting next year, California’s Senate Bill 1383 will begin to require businesses and consumers to separate food waste from their garbage and put them into “green” bins for proper composting.

The legislation, passed in 2016 by then-governor Jerry Brown and the California state legislature, also will begin requiring tier 1 food businesses (grocery retailers, food distributors, food service providers) to divert 20% of food destined to be thrown away to food recovery organizations by the year 2025.

While I think it’s a good thing that everyone in California – both consumers and businesses – will eventually be required to start composting, I’m more interested in how SB 1383 could catalyze interest in platforms that help put food destined for the waste bin on someone’s plate instead. After all, while composting is a net positive from a climate impact reduction perspective, it’s also the last stop on the food waste recovery and mitigation express. In other words, when food feeds someone instead of ending up as fertilizer, everyone wins.

Image Source: CalRecycle SB 1383 Overview

The timing is good for the law, partly because the pandemic has driven home the realization among businesses that it’s their responsibility to try and divert food to local food banks or other food recovery organizations as good corporate citizens. And of course, it also makes good business sense, since by redirecting food to food recovery organizations, these businesses can also claim these donations on their taxes.

As grocery retailers and other tier 1 food businesses ramp up their food diversion efforts, there are some organizations that could help them along the way. One of these companies is Goodr, which offers grocery retailers and other food-related organizations a tech platform and associated service to help them get excess food in the hands of food charities. Goodr sprang into action in its home market of Atlanta during the early days of the pandemic and showed it could really make an impact. Other organizations such as Quest also provide food diversion services and food waste audits.

One of the challenges of a food recovery program is just having the ability to track and manage potential food waste. There are a number of technology platform providers such as Afresh and Crisp that give grocery providers tools powered by machine vision, AI, and other cutting-edge technologies to better predict and manage fresh food inventories. There are even food robot companies like Simbe developing technologies to help assist in food waste reduction management.

Finally, there are marketplaces like Olio and Too Good to Go that enable grocery retailers, restaurants, and other organizations to list excess edible food for sale on a highly-discounted basis to local consumers. While food sold on these platforms will not count towards the company’s 20% food diversion requirement, using them will help a company reduce the overall amount of food wasted and help provide low-cost food to consumers.

But what I am most excited about is how SB 1383 could give rise to new solutions to help food retailers and foodservice providers waste less food. New regulations often serve as catalysts for innovation, giving large businesses a new reason to invest in core technology infrastructure. As SB 1383’s regulations begin to go into effect, innovators with good ideas for new technology to help companies reduce food waste and redirect excess food towards food insecure citizens will have a growing market opportunity for their solutions. This growing opportunity will likely attract more venture investment for a category that has, at least in the past, had a hard time convincing investors there was enough of a market to garner them a return on their investment.

FloWaste Raises a $1.1M Pre-Seed Round To Reduce Food Waste With Machine Learning

According to the Environmental Protection Agency, more than 85% of greenhouse gas emissions from landfilled food waste can be attributed to missteps that occur before the food ever reaches a consumer’s plate—from production to processing to distribution. FloWaste, an Indiana-based startup, is addressing food system inefficiency at the processing stage using a proprietary machine learning system.

FloWaste announced today that it has raised a $1.1 pre-seed funding round, which it will use to scale up and improve its technology. Last week, The Spoon got on Zoom with company founder and CEO Rian Mc Donnell to find out how FloWaste can help food producers send less food to landfills.

Mc Donnell had the idea for FloWaste while studying mechanical and manufacturing engineering at Trinity College Dublin. “I gravitated toward the topic of food waste because by that point, I knew that whatever I did, my life was going to be sustainability-focused,” says Mc Donnell.

Here’s how FloWaste works: Customers identify 20 foods that they’d like the system to track, and the team trains their machine learning system to recognize those foods. Then the team installs cameras above customers’ workstations, production lines, and trash cans. The cameras monitor the food production process, automatically classifying food items and quantifying how much gets thrown away.

Flowaste Animation

Video: FloWaste food identification and qualification. Source: FloWaste

“We gather a ton of data,” says Mc Donnell. “And we can chop and change that data based on ingredient usage, yield, shift performance, or daily performance. We present those insights to the management, and then they can make procedural changes.”

The technology can be used in both the industrial and the commercial food sectors. One of FloWaste’s current customers, a European protein producer, is using the system to monitor waste on a beef production line. “There’s this huge financial return because proteins are expensive,” says Mc Donnell, “but also this huge environmental return because any increase in yield means that you’re effectively killing less cows in the long term.”

According to Mc Donnell, the task of training the machine learning system to recognize different foods has been time-consuming. But he hopes that as the system builds knowledge, it’ll become easier and easier to expand its use. “If we go in with someone and they’re doing fries, it means the next time we go and search for fries, we’ve already got a head start,” he says. “We’re slowly getting more and more robots to the point where eventually we’ll be able to just do a general use case of food as a whole.”

This pre-seed funding round will help FloWaste to build up scale with its technology: The company has signed agreements to launch the system at over 100 locations with its pilot customers in the next nine months. The funding will also help the team contend with the challenges of creating a hardware system from scratch using off-the-shelf cameras. In the near future, Mc Donnell is planning to bring on a full-time IoT engineer to make the system simpler and more reliable.

FloWaste is participating in the current cohort of Europe’s Rockstart accelerator-VC. The company has also received funding from U.S. venture funds, including Underdog Labs and Flywheel Fund.

In the longer term, the team hopes to expand through new partnerships. “We’re working on installing in cafeteria kitchens and doing post-consumer analytics for customers who want that,” says Mc Donnell. “And we’re looking at quick-service restaurants because they have such an emphasis on optimizing their processes and their yield of food in the kitchen.”

Ultimately, the company is on a mission to help food producers discover how more environmentally friendly processes can also boost margins. “The best way to see a sustainability benefit is to tie it to the financials of business,” says Mc Donnell, “and actually teach businesses how they can be making more money by being more sustainable.”

Soft Robotics Wants To Give Your Food Robot Good Eye-Hand Coordination

Football wide receivers that can catch the ball well are said to have soft hands. Food robots who use too much force grabbing delicate produce are, well, just being robots.

But now, robot system designers can turn their food robot into a veggie-shuffling Jerry Rice with the new mGrip “hand” from Soft Robotics.

The mGrip is part of a new SoftAI product suite from Soft Robotics that robot designers can add to existing systems to optimize them for handling food like meat and produce in high-volume environments. In addition to a food-grabbing hand, the SoftAI suite includes a “perception module” that pairs cameras with machine vision software that the company says will add “eye-hand coordination” to industrial robots. The on-board processing of the perception module uses machine learning to understand how to categorize and segment different types of food.

Robotics has long been used for tasks like packing food, but only in highly structured environments. Often, this meant using humans as part of the process to do things like sorting. However, advances in machine vision over the past couple of years have meant machines can essentially replace the need for humans to do some of the tasks they’ve been needed for in the past. With an off-the-shelf product like SoftAI, what companies like Amazon have probably spent millions to build now becomes more turnkey.

Beyond high-volume warehousing applications, I can also see how a platform like SoftAI could be used in more consumer-facing food robotic systems. For example, first-generation robotic food kiosks often used off-the-shelf robotic arms that required lots of customization to make them work. With new food-optimized plug-in hardware like SoftAI, small teams could accelerate their time to market and dedicate their time to other engineering problems.

You can see the mGrip and the SoftAI perception module in the video below.

Foodspace is Using AI to Create Better CPG Data So You Find That Spicy Cheese Faster

You ever search for a food product online or at the grocery store but can’t find that exact something that perfectly matches your taste, dietary or nutritional preferences? You’re not alone. One of the big reasons searching for food products can be so frustrating is they are often bucketed under data categories that are holdovers from existing category management systems built fifteen or twenty years ago.

A new startup called Foodspace wants to eliminate this annoying experience by helping the CPG and food retail industry update their old-school category management systems with technology that makes sure that every conceivable product attribute a consumer may be searching for is documented and assigned to products headed to a physical or digital shelf.

The Boston-based startup plans on doing that by using machine vision technology that analyzes scanned images of new product packaging introduced by CPG manufacturers and uses AI to synthesize and assign attributes based on its understanding of the product packaging and label data. The attributes go beyond the typical high-level product categories such as organic or gluten free, and factors in things such as sensory preferences (creamy, grainy, etc) and consumer taste and lifestyle archetypes. All told, Foodspace’s system can assign nearly three thousand different attributes to a product.

The end result should be faster, more personalized searches for consumers. If, for example, a person who likes cheese, loves spicy food, and has a gluten allergy heads to the deli section of an online grocery store, they shouldn’t have to drill down five categories deep within the deli category to find that gluten-free habanero cheddar. With Foodspace’s AI-powered synthesis and matching of different attributes, a consumer finds a product match much faster, perhaps almost immediately, depending on the understanding the online grocer has about the shopper.

Of course, this move towards more granular, highly-consumer centric data is something that CPG and retail industries recognize is important, but have been slow to evolve away from because of the huge magnitude of switching towards systems that have thousands of product attributes. The Food Industry Association (which goes under the acronym FMI), has been working on a new framework called Shopper Centric Retailing that would update product information in the more detailed way, and this week at FMI’s annual midwinter meeting, the industry consultant who developed Shopper Centric Retailing framework, Winston Weber, announced Foodspace as a “premier” strategic solution partner to help food product companies transition their products to the new format.

In short, Weber sees Foodspace’s technology as an enabling platform to help food brands migrate to the future.

Foodspace’s technology is “helping translate products in the online space, to the benefit of brands, retailers and the end consumers,” said Weber CEO and namesake Win Weber in the press release. “Their technology is the conduit for which the Shopper-Centric Retailing business model can optimize consumer satisfaction.”

As I thought about better product data that could personalize my food product searches, I started to wonder if this could help usher in the personalized food profile concept that I’ve been thinking about ever since I heard Mike Lee talk about the idea at Smart Kitchen Summit in 2017.

Foodspace CEO Ayo Oshinaike thinks so. “The universal data set that enables that is not there,” Oshinaike told me via Zoom. “That’s the piece that’s in the middle that Foodspace is trying to solve with the breaking down of the information accuracy and how we’re able to relate products to consumers.”

Is Nest’s New Face Recognition Cam A Sign It’s Waking From Its Slumber?

One of the great mysteries of the smart home world over the past few years has been the relative quiet of Nest, the one-time connected home star that burst onto the scene with the launch of its impressive Nest Learning Thermostat and, two years later, a smoke and carbon monoxide detector by the name of Nest Protect.

After Nest got acquired in early 2014, updates slowed to a crawl, and much of the news that did come out from the company during this time was bad. While there was occasional news about Works with Nest and Thread, you were just as likely to read about management dysfunction under Tony Fadell and product recalls. The only wholly new product line introduced into the Nest family during this time was the Nest Cam, a product that, in reality, owes more to Dropcam (another acquisition by Google) than to internal development from Nest.

But now there are indications the company might be waking up from its long slumber. In a recent story in The Verge, Nest product manager Maxine Verson hinted that the rest of 2017 should be busy for the company:

Verge writer Vlad Savov writes, “Veron tells me Nest’s relative silence in recent times is about to be a thing of the past. “I am very excited about the next six months,” he says with a grin, “we’ll talk again soon and you’ll understand why.”

Savov goes on the speculate that the next product might be a cheaper home thermostat. A welcome addition to be sure, but I think for those who witnessed Nest’s early days of innovation, a lower-cost version of an existing product is hardly something to get excited about.

The product I’m excited to see is Nest’s long-rumored home security system. My own sources have confirmed the existence of this long-gestating project and, given Google’s patent filings in the space, I think a Nest home security system could be truly differentiated. Add in the fact they just introduced a camera with facial recognition capabilities – an interesting potential component of a smart security system – and we may be getting close.

Another potential product is a video doorbell.  While the market is certainly crowded at this point, Nest’s brand name and recent development of a Pro group that supports home builders and integrators could help a Nest video doorbell get traction.

Whatever Nest does, chances are it’s been limited to a certain set of products by its parent company. Alphabet/Google has let non-Nest groups develop products in some of the most interesting areas – voice assistants and mesh Wi-Fi for example – while Nest has largely stuck with thermostats and cameras.

One thing is certain: the company’s new Nest Cam itself is a sign of progress. With it, the company has started to integrate image-based AI into its Nest cams, a potential indication that it – like Amazon – sees computer vision as one of the key new frontiers in the smart home.

And who knows? Maybe now – if the new camera and hints being dropped are any indication – maybe Nest truly has something new and interesting up its sleeves.

Make sure to subscribe to the Spoon newsletter to get it in your inbox. And don’t forget to check out Smart Kitchen Summit, the only event on the future of the food, cooking, and the kitchen. 

Maker Of A Much-Changed SmartPlate Hopes To Ship This Summer

When the Sharks bite, sometimes you just have to pivot. Or, at the very least, tweak your hardware design a bit.

That’s what happened with Fitly and their product SmartPlate.

The original SmartPlate, which managed to raise $110 thousand on Kickstarter, included three embedded cameras, a weight sensor, and wireless connectivity. The product worked like this: the plate’s embedded cameras would analyze the food using image recognition software and, combined with the embedded weight sensor, spit out a nutritional analysis in the associated smartphone app.

If that sounds like an overly complicated and expensive solution, that’s because it is. And when Fitly employee Martin Dell’Arciprete appeared on Shark Tank in the fall of 2015 (the show aired in February 2016) to pitch the SmartPlate, the Sharks were quick to smell blood in the water. Cuban and company echoed what many others had thought: why would someone buy a $200 plate with embedded cameras and sensors when you could do much of the same with a phone app?

Kick Calorie Counting with SmartPlate - Shark Tank

It was a brutal showing. Not long after the appearance, Dell’Arciprete was fired by company CEO Anthony Ortiz. A partnership with former Olympian Jackie Joyner-Kersee, announced about the time the company was going on Shark Tank, was never mentioned again.

But less than a year after the episode aired, Ortiz and the SmartPlate were back raising funds, this time with a much different product. Now called the SmartPlate Topview, the product was now modular and included a base with all the embedded electronics and separate plastic plate that looked much like the previous tri-portion plate, but only this time without any embedded electronics. Gone were all of the cameras, as all picture taking duties and image analysis were now performed by the smartphone and the SmartPlate app.

For the new campaign, Ortiz and company moved the revamped SmartPlate to Indiegogo. The pitch was much the same, including some of same original video clips and backstory as told by Ortiz on the Kickstarter campaign. However, the price for the new SmartPlate were also lower this time around, with estimated retail for the revamped product at $149 (with three plates and a base), compared to the $199 for the original SmartPlate.

Most would look at this changes and say they were necessary; after all, one has to wonder why a plate would ever need its own camera, let alone three.

Still, even now, I am not sure why an actual plate system is necessary. While the company states that the base and associated plates together give the best overall performance, they admit that you can use the app by itself to analyze food without the plate.

But after a long journey which included two crowdfunding campaigns, the company looks like they may accomplish what looked pretty impossible just a year ago: shipping product. According to a recent update by Ortiz on their Indiegogo page, Fitly has ordered 10 thousand units for production.

And sure, with all the bumps in the road, it’s easy to be skeptical. But the company has someone vouching for them this time. According to an Arrow spokesperson, the new SmartPlate’s electronics and software are ready for manufacturing. My guess is that Arrow probably wouldn’t put their name on this campaign if they didn’t think it could ship, so it looks like backers may actually soon getting their SmartPlates.

Want a sneak preview of SmartPlate? You can download the app and check it out. And who knows? Maybe you’ll like the app so much you’ll buy the plate.

More Than Hot Dogs: Pinterest & Google Image Recognition AI Make A ‘Shazam For Food’ Possible

In this season of Silicon Valley, one story line has housemate and programmer Jian-Yang developing a food recognition app called ‘See Food.’ Since the idea was born out of a spitballed pitch for a “Shazam for Food” by Yang’s landlord Erlich Bachman, it’s not altogether surprising that when Jian-Yang finally gets around to hacking together the app, it’s only good for one thing: telling us whether whatever is in front of the camera is a hot dog or not a hot dog (and yes, the once-fake app is now a real fake app you can now download for real in the app store).

Silicon Valley: Not Hotdog (Season 4 Episode 4 Clip) | HBO

While a “Shazam for Food” pitch seems like the perfect sendup concept for a satirical show about the tech world, the truth is there has been significant advances in machine vision and learning in the past few years that make food recognition a very real and potentially useful application.

These advances were on display this month by both Google and Pinterest as they both touted image recognition services called “Lens”.  While Pinterest has been working on image search since at least 2015, they rolled out their Lens this past February. In a blog post, Pinterest CEO Evan Sharp highlighted a food use-case as an example of how Lens could work.

“You can also use Lens with food,” wrote Sharp. “Just point it at broccoli or a pomegranate to see what recipes come up.”

And this week, after Google launched a similar feature with the same name, Pinterest apparently felt they needed to emphasize the food recognition capability of their Lens offering in the form of a new blog post that repeated what Sharp told us in February: users can use Lens to serve up recipe suggestions.

“Our visual discovery technology already recognizes objects in more than 750 categories, and people have been busily pointing Lens beta at everything from lemons to strawberries to find new recipes to try. And now we’re rolling out a way for you to Lens an entire dish and get recipes to recreate the meal.”

Google introduced its Lens image recognition technology last week at its annual developer conference, Google I/O. Not that Google is new to image AI or even food recognition. The company has been working on image search for probably close to a decade, and in 2015 introduced an app called Im2Calories that gave calorie estimates of food based on image analysis of food.  And while Google didn’t highlight any specific food use cases for their version of Lens at I/O, there’s no doubt that the company and its partners will explore using image AI to surface information such as recommendations for recipes like Pinterest or nutritional information (like Im2calories).

Of course, all of this follows Amazon’s recent push into image AI with the debut of its own camera-enabled Echo devices and the continued maturation of its AWS AI service, Rekognition. My guess is that as big players double down on voice assistant services, image recognition and analysis has reached a maturation point that makes it ready for consumer applications.

While clumsy efforts at food recognition like the Smart Plate  – as well as two big companies launching an image recognition service with the same exact name – make food image recognition a ready topic for satire, the reality is the technology is reaching a point of maturity and usefulness that maybe – just maybe – we’ll soon have a Shazam for Food that consumers really want.