Hi Guys,
I am currently investigating the AI builder models en trying to find use-cases for this.
I currently work at a cardboard manufacturer and one of the struggles they have is classifying 'cloudiness' of cardboard. This is to say the flatness or evenness of the cardboard.
Pictures are taken from samples with a camera straight from the top, with a flash. the amount of light-'particles' visible will determine the classification is. The more visible light particles, the worse the grade of the paper, as more reflections show there is more unevenness.
My question:
Can you teach the Object detection model to recognize tens, if not hundreds of of the same objects (light particles in this case) in one picture or will this not be viable with these models?
Hope to hear what you think!
Regards,
Peter
Solved! Go to Solution.
Hi Peter!
Jake here, a PM on the AI Builder team. Are you looking for the model to provide the # of light particles in a given image of cardboard? Or to just classify if an image of cardboard is flat vs not, or on a scale of flat, somewhat flat, very much not flat?
If it's the latter, this sounds like a great fit for Image Classification. Instead of labeling boxes around each light reflection, you can label images as "flat" "somewhat flat" and "very much not flat" given the # of light particles in the entire image. This way the model starts to learn that the amount of light reflection is what classifies an image as flat vs somewhat flat vs very much not flat.
I would recommend using Lobe (https://lobe.ai), a free easy-to-use desktop app by Microsoft to train custom machine learning models. If you can download these images of cardboard to your desktop, you can easily import and label the images and it will automatically train a model. I would recommend starting with a hundred images in those 3 groups to get a proof of concept and see how it performs.
If you'd like to chat more about your business scenario or if you have more questions about Lobe, feel free to email us aihelpen@microsoft.com.
Hi Peter!
Jake here, a PM on the AI Builder team. Are you looking for the model to provide the # of light particles in a given image of cardboard? Or to just classify if an image of cardboard is flat vs not, or on a scale of flat, somewhat flat, very much not flat?
If it's the latter, this sounds like a great fit for Image Classification. Instead of labeling boxes around each light reflection, you can label images as "flat" "somewhat flat" and "very much not flat" given the # of light particles in the entire image. This way the model starts to learn that the amount of light reflection is what classifies an image as flat vs somewhat flat vs very much not flat.
I would recommend using Lobe (https://lobe.ai), a free easy-to-use desktop app by Microsoft to train custom machine learning models. If you can download these images of cardboard to your desktop, you can easily import and label the images and it will automatically train a model. I would recommend starting with a hundred images in those 3 groups to get a proof of concept and see how it performs.
If you'd like to chat more about your business scenario or if you have more questions about Lobe, feel free to email us aihelpen@microsoft.com.
Hi @cohenjacobd ,
Thank you for your reply. I thought about it a bit more, and i thought of a better analogy of what we're trying to accomplish.
Imagine snow on the tv-screen. Black and white particles, imagine that the screen would be fully black without the snow and the snow particles are white.
This directly translates to my use-case as the images to be measured are also black and white. If there is no snow in the image, this indicates the surface is smooth and the quality is perfect.
Every snowflake that is on the image should be counted. This way, we can create a ranking similar to":
#1: 0 - 10 snowflakes,
#2: 10 - 50 snowflakes
#3: 50 - 100 snowflakes
and so on...
I would add some images here, but have yet to receive one sadly.
I am going to try import some images in Lobe as soon as i receive them and see if it will help us! Thanks for taking the time to respond, much appreciated!
Regards,
Thanks for sharing more details. Is the intent for the model to categorize "good" vs "bad" cardboard? Like the good cardboard goes off the manufacturing line for sale, and the bad cardboard is recycled or used for something else, or sold for less? Or is this to alert the manufacturing line for anomalies when the occur? Or just overall analysis to assess how well the manufacturing of cardboard is going over time?
Please do let us know how it goes training a model with Lobe. If having 3 categories based of low, medium, high amount of snowflakes is sufficient for your needs, then I think image classification is a good starting point.
Jake
Hi @cohenjacobd ,
The intent is for our R&D department when developing new types of board and to increase quality of existing board. It's a way of objectifying the roughness of the board. Up untill now it is only measured in the eye of the beholder.
I was expecting to be able to import the Lobe model in the Power Platform AI model, but it doesn't seem this is a possibility? This means that i would have to build my own application around it. I have some developement skills, but have to see if i can tackle this.
I did some testing and it seems the model works really well! As i haven't received any testmaterial yet, I did some experiments on my own, and it seems it correctly predicts what grade it belongs in, 97% of the time (Grade 1 - 4 in my case, where 1 is bad and 4 is really well).
Next step is actual testing, have to wait for that until next week.
Thanks for your help so far!
Regards,
Great to hear your testing so far is working well! Excited to hear how the test material works next week.
We are actively developing the support to import a Lobe model to Power Platform, and will start a private preview soon 🙂 Would you like to join the private preview? Our team can provide assistance in building your business solution.
Jake
I would love to join the private preview!
I have a couple of years of experience with building apps and flows so that should not be an issue.
Let me know if you need anything for me to join the preview.
Thank you for the invitation!
Regards,
Peter
Hi Peter! I have sent you a personal message inviting you to the private preview. Please check that out and sign up and we can follow up more. I'm looking forward to hearing more about your progress 🙂
Jake
Dear Community Members, We'd like to let you know of an upcoming change to the community platform: starting July 16th, the platform will transition to a READ ONLY mode until July 22nd. During this period, members will not be able to Kudo, Comment, or Reply to any posts. On July 22nd, please be on the lookout for a message sent to the email address registered on your community profile. This email is crucial as it will contain your unique code and link to register for the new platform encompassing all of the communities. What to Expect in the New Community: A more unified experience where all products, including Power Apps, Power Automate, Copilot Studio, and Power Pages, will be accessible from one community.Community Blogs that you can syndicate and link to for automatic updates. We appreciate your understanding and cooperation during this transition. Stay tuned for the exciting new features and a seamless community experience ahead!
We are excited to announce the Summer of Solutions Challenge! This challenge is kicking off on Monday, June 17th and will run for (4) weeks. The challenge is open to all Power Platform (Power Apps, Power Automate, Copilot Studio & Power Pages) community members. We invite you to participate in a quest to provide solutions in the Forums to as many questions as you can. Answers can be provided in all the communities. Entry Period: This Challenge will consist of four weekly Entry Periods as follows (each an “Entry Period”) - 12:00 a.m. PT on June 17, 2024 – 11:59 p.m. PT on June 23, 2024 - 12:00 a.m. PT on June 24, 2024 – 11:59 p.m. PT on June 30, 2024 - 12:00 a.m. PT on July 1, 2024 – 11:59 p.m. PT on July 7, 2024 - 12:00 a.m. PT on July 8, 2024 – 11:59 p.m. PT on July 14, 2024 Entries will be eligible for the Entry Period in which they are received and will not carryover to subsequent weekly entry periods. You must enter into each weekly Entry Period separately. How to Enter: We invite you to participate in a quest to provide "Accepted Solutions" to as many questions as you can. Answers can be provided in all the communities. Users must provide a solution which can be an “Accepted Solution” in the Forums in all of the communities and there are no limits to the number of “Accepted Solutions” that a member can provide for entries in this challenge, but each entry must be substantially unique and different. Winner Selection and Prizes: At the end of each week, we will list the top ten (10) Community users which will consist of: 5 Community Members & 5 Super Users and they will advance to the final drawing. We will post each week in the News & Announcements the top 10 Solution providers. At the end of the challenge, we will add all of the top 10 weekly names and enter them into a random drawing. Then we will randomly select ten (10) winners (5 Community Members & 5 Super Users) from among all eligible entrants received across all weekly Entry Periods to receive the prize listed below. If a winner declines, we will draw again at random for the next winner. A user will only be able to win once overall. If they are drawn multiple times, another user will be drawn at random. Individuals will be contacted before the announcement with the opportunity to claim or deny the prize. Once all of the winners have been notified, we will post in the News & Announcements of each community with the list of winners. Each winner will receive one (1) Pass to the Power Platform Conference in Las Vegas, Sep. 18-20, 2024 ($1800 value). NOTE: Prize is for conference attendance only and any other costs such as airfare, lodging, transportation, and food are the sole responsibility of the winner. Tickets are not transferable to any other party or to next year’s event. ** PLEASE SEE THE ATTACHED RULES for this CHALLENGE** Week 1 Results: Congratulations to the Week 1 qualifiers, you are being entered in the random drawing that will take place at the end of the challenge. Community MembersNumber of SolutionsSuper UsersNumber of Solutions @anandm08 23 @WarrenBelz 31 @DBO_DV 10 @Amik 19 AmínAA 6 @mmbr1606 12 @rzuber 4 @happyume 7 @Giraldoj 3@ANB 6 (tie) @SpongYe 6 (tie) Week 2 Results: Congratulations to the Week 2 qualifiers, you are being entered in the random drawing that will take place at the end of the challenge. Community MembersSolutionsSuper UsersSolutions @anandm08 10@WarrenBelz 25 @DBO_DV 6@mmbr1606 14 @AmínAA 4 @Amik 12 @royg 3 @ANB 10 @AllanDeCastro 2 @SunilPashikanti 5 @Michaelfp 2 @FLMike 5 @eduardo_izzo 2 Meekou 2 @rzuber 2 @Velegandla 2 @PowerPlatform-P 2 @Micaiah 2 Week 3 Results: Congratulations to the Week 3 qualifiers, you are being entered in the random drawing that will take place at the end of the challenge. Week 3:Community MembersSolutionsSuper UsersSolutionsPower Apps anandm0861WarrenBelz86DBO_DV25Amik66Michaelfp13mmbr160647Giraldoj13FLMike31AmínAA13SpongYe27 Week 4 Results: Congratulations to the Week 4 qualifiers, you are being entered in the random drawing that will take place at the end of the challenge. Week 4:Community MembersSolutionsSuper UsersSolutionsPower Apps DBO-DV21WarranBelz26Giraldoj7mmbr160618Muzammmil_0695067Amik14samfawzi_acml6FLMike12tzuber6ANB8 SunilPashikanti8
On July 16, 2024, we published the 2024 release wave 2 plans for Microsoft Dynamics 365 and Microsoft Power Platform. These plans are a compilation of the new capabilities planned to be released between October 2024 to March 2025. This release introduces a wealth of new features designed to enhance customer understanding and improve overall user experience, showcasing our dedication to driving digital transformation for our customers and partners. The upcoming wave is centered around utilizing advanced AI and Microsoft Copilot technologies to enhance user productivity and streamline operations across diverse business applications. These enhancements include intelligent automation, AI-powered insights, and immersive user experiences that are designed to break down barriers between data, insights, and individuals. Watch a summary of the release highlights. Discover the latest features that empower organizations to operate more efficiently and adaptively. From AI-driven sales insights and customer service enhancements to predictive analytics in supply chain management and autonomous financial processes, the new capabilities enable businesses to proactively address challenges and capitalize on opportunities.
We're embarking on a journey to enhance your experience by transitioning to a new community platform. Our team has been diligently working to create a fresh community site, leveraging the very Dynamics 365 and Power Platform tools our community advocates for. We started this journey with transitioning Copilot Studio forums and blogs in June. The move marks the beginning of a new chapter, and we're eager for you to be a part of it. The rest of the Power Platform product sites will be moving over this summer. Stay tuned for more updates as we get closer to the launch. We can't wait to welcome you to our new community space, designed with you in mind. Let's connect, learn, and grow together. Here's to new beginnings and endless possibilities! If you have any questions, observations or concerns throughout this process please go to https://aka.ms/PPCommSupport. To stay up to date on the latest details of this migration and other important Community updates subscribe to our News and Announcements forums: Copilot Studio, Power Apps, Power Automate, Power Pages