FruitPunch AI

🔔🔔 Last chance to sign up for the AI for Indian Farmer: Handwritten Devanagari Challenge! 🔔🔔 We are kicking off this 10-week Challenge on the 21st of October to develop Optical Character Recognition for handwritten Devanagari script to close the digital divide for Indian Farmers 🇮🇳 #OCR has been solved for western/Latin script for over 20 years, yet it is still too inaccurate to use for handwritten Devanagari script. This keeps in place a huge digital divide, disallowing Indian farmers the same benefits of western farmers like yield prediction and loan calculation. The AI for Indian Farmer: Handwritten Devanagari Challenge aims to take first steps in solving this problem for good together with Heifer International & Heifer India 🐄 We are looking for machine learning engineers who want to use their skills to make a positive impact. In this Challenge, you will work with an international team to scope this problem, scout for datasets and develop new OCR models. We will explore traditional machine and deep learning models to determine the most effective approaches. Among others, the following topics will be covered in the challenge: - Computer Vision (e.g. OpenCV, PyTorch, TensorFlow, etc.) - Deep Learning Methodologies Convolutional Neural Networks Generative - Adversarial Networks - Recurrent Neural Networks - Feature Extraction Techniques - NLP / LLM use This is a perfect project to up-skill yourself in the field of NLP and CV! 🧑🎓 Sign up now: https://lnkd.in/exEpF9q4 Vess Antoinette Buster Dorian #devanagari #LLMs #digitaldivide #agritech

After failing to solve OCR for handwritten Devanagari script in two of our Challenges (!) we realized this has been an unsolved ML problem for over two decades. This article outlines what we've tried so far and why it's especially hard for Devanagari script as opposed to Latin scripts for example. Now we're going to give it a go with more data from Heifer International India 😊 Likely this will require multiple Challenges to solve, but once we do, we'll be able to bring the advantages of AI analysis to Northern Indian farmers and many other nations in the region using Devanagari. #OCR #MachineLearning #digitaldivide #ai4good #india #agritech Marcus Dorian Buster Karan

OCR* has been solved for western/Latin script for over 20 years, yet it is still too inaccurate to use for handwritten Devanagari script. This keeps in place a huge digital divide, disallowing Indian farmers the same benefits of western farmers like yield prediction, loan calculation etc. This is a huge problem!  So we’re launching a new Challenge with Heifer International: 🇮🇳 AI for Indian Farmers - Handwritten Devanagari ✍️ 🎯The goal: Develop Optical Character Recognition for handwritten Devanagari script to close the digital divide for Indian Farmers 💡 So why is this old Machine Learning problem still unsolved? While the Latin script has 26 letters and a total of ~70 different characters incl. numbers and punctuation, Devanagari has thousands of combinations due to the conjunction of consonants, diacritics and modifiers. As a cherry on the cake, letters are joined together in a word as well. Heifer Labs & Heifer India have provided us with big datasets from the handwritten financial records of farmers to start solving this problem. We’re looking for Experts with to help solve this Challenge, apply & more info here: https://lnkd.in/exEpF9q4  ❣️ Big plus if you can read Devanagari ❣️ Technical objectives: 🖼️ Image Preprocessing: Improve the quality of the images so that the OCR solutions would be standardised/improved. Identify which techniques improve the quality of OCR #️⃣ OCR for Devanagari Script Develop an OCR solution to extract both handwritten and typeset text from loan documents. The OCR system must handle the complexities of the Devanagari script effectively. ✅ Post-processing / Quality Control Assess the quality of the OCR outputs Make a best guess on any uninterpretable characters and determine if the spelling of words is correct. Topics that will be covered: #ComputerVision (e.g. OpenCV, PyTorch, TensorFlow, etc.) #DeepLearning Methodologies #CNN #GAN (Generative Adversarial Networks) #RNN (Recurrent Neural Networks) #FeatureExtraction Techniques and #NLP *#OCR: Optical Character Recognition: computer vision technique to turn images of letters and numbers into their digital equivalents Vess Antoinette Buster Kishore Tanishka Juhee Reena Deepa #devanagari #LLMs #digitaldivide #agritech

🔔🔔 Last chance to sign up for the AI against Infant Death Challenge! 🔔🔔 We are kicking off this 10-week Challenge on the 1st of October to develop machine learning models for life-saving medical devices currently in use in Malawi 🇲🇼. The AI against Infant Death Challenge aims to accurately assess the risk and detection of deterioration in hospitalized babies and children in Africa. Using big Data and Machine Learning we will look for early warning signs. And you can help! 🧑⚕️ We are looking for machine learning engineers who want to use their skills to make a positive impact. In this challenge, you will work with an international team to develop new AI models to classify diagnoses and predict the progression of young patients. We will explore traditional machine and deep learning models to determine the most effective approaches. Among others, the following topics will be covered in the challenge: - Signal processing - Forecasting - Classification - Explainable AI This is a perfect project to up-skill yourself in the field of AI and Healthcare! 🧑🎓 Sign up now: https://lnkd.in/eSTjC3fd

Not even having the time to resuscitate a child... That's what we're helping to prevent with our new Challenge: AI against Infant Death. Niek, CEO of GOAL 3 the organization behind this Challenge shares a documentary about the work they're doing and that we'll be helping out in by building ML models 🙏 If you want to help out, sign up here: https://lnkd.in/eSTjC3fd #aiforhealth #goal3 #SDG3 Agnes Arnoud

Each year nearly 5 million infants die *mostly preventable deaths* in Africa. We will be working with GOAL 3 to improve this terrible statistic in our next Challenge: AI against Infant Death ❤️🩹 🙏 how? 👇 🎯The goal: Develop ML models for their life-saving medical devices used right now in Malawi 🇲🇼 To accurately assess the risk of and detect deterioration in hospitalized babies & children in Africa, a wealth of data has been collected from infants connected to GOAL 3’s IMPALA System. Right now the medical “ICLEN" score is used to estimate the health status which only uses limited parameters about the infant like heart rate, consciousness and respiratory rate, whereas IMPALA has a lot more data to go on 🗄️ Join this Challenge and you’ll work directly together with doctors and data scientists, using machine and deep learning to: 1️⃣ Develop models to predict deterioration of the infant 2️⃣ Use data from ballisto(cardio)graphic sensors and correlate it to data from other sensors (like ECG) and clinical data (like consciousness) 3️⃣ Implement explainable AI: determine which variables influence the risk of death the most The data: 170M seconds of data with vital signs and traces. Traces is about 12TB of data. 55.000 unique hours of data with summarized data per hour 🤯 this data is provided by Vrije Universiteit Amsterdam (VU Amsterdam) as part of the IMPALA consortium 🦌 a collaboration between African and European Universities. 💪 Skills relevant for this project are: #Signalprocessing, #Regression, #Forecasting, #Classification & #Whiteboxmodeling Next to Goal 3 we’re proud and humbled to work with engineers from Achmea and the Achmea Foundation as sponsor of this Challenge ❤️ Apply & more info here: https://lnkd.in/eSTjC3fd You can apply as an Expert or join as an Apprentice - then you’ll get Mentorship as well 😊 #aiforgood #ai4good #aiforhealth #healthcare #sdg3 Niek Agnes Jelle Dorian Job Daan Arnoud Bart Erwin Arjan Lonneke

🇳🇵 यदि तपाईंले यो पढ्न सक्नुहुन्छ भने हामीलाई तपाईंको मद्दत चाहिन्छ 🇳🇵 (If you can read this we need your help 😉) 🔬 For our upcoming AI for Women Farmers Challenge we are looking for AI for Good Engineers with a very specific profile! 🇳🇵 In collaboration with Heifer International, we will digitize handwritten Nepali databases using AI. Specifically, we will be addressing Optical Character Recognition for Devanagari script, which is the script used for Nepali and some of the most spoken Indian languages such as Hindi. We are doing this to assist agricultural cooperatives in making better use of their data. We would greatly benefit from Nepali-speaking data scientists with expertise in OCR. 🦄🦄 We're aware that those are rare. But if you meet 𝐚𝐧𝐲 𝐨𝐧𝐞 of these requirements we can really use your help: - Nepali native / or roots - Proficient in reading Devanagari script - Experience with OCR for handwritten text The Challenge starts on the 2nd of July and will take 10 weeks to complete! Check out the Challenge on our platform and "Apply as an expert" for free if you meet one of the requirements. https://lnkd.in/eiBe8hKj If you know anyone who meets these requirements please connect them to our Challenge manager Dorian Groen Applications are officially open until the 30th of June, but we can make an exception if you can valuably contribute to the Challenge 😁 #AIforGood #AIforAgriculture #DataScience #SDG

🇳🇵🌾 𝐋𝐚𝐬𝐭 𝐜𝐚𝐥𝐥 𝐭𝐨 𝐣𝐨𝐢𝐧 𝐭𝐡𝐞 𝐀𝐈 𝐟𝐨𝐫 𝐖𝐨𝐦𝐞𝐧 𝐅𝐚𝐫𝐦𝐞𝐫𝐬 2 𝐂𝐡𝐚𝐥𝐥𝐞𝐧𝐠𝐞! 🌾🇳🇵 You can use your expertise in this Challenge designed to enhance agricultural practices in Nepal, benefiting both individual farmers and cooperatives. 🔍 𝐂𝐨𝐧𝐭𝐞𝐱𝐭 𝐨𝐟 𝐭𝐡𝐞 𝐏𝐫𝐨𝐛𝐥𝐞𝐦: In Nepal, agricultural data management faces significant hurdles due to its dependence on paper records. This challenge hinders farmers' and cooperatives' ability to make informed decisions, affecting their financial and agricultural productivity. Our goal is to digitize this system using computer vision and AI, offering digital solutions for a brighter, more sustainable future in agriculture. 𝐖𝐡𝐲 𝐄𝐧𝐠𝐚𝐠𝐞 𝐚𝐬 𝐚𝐧 𝐄𝐱𝐩𝐞𝐫𝐭? 𝐑𝐞𝐚𝐥-𝐖𝐨𝐫𝐥𝐝 𝐈𝐦𝐩𝐚𝐜𝐭: Make a tangible difference by translating and analyzing farmer loan data, which will lead to enhanced agricultural operations, improved financial strategies, and more sustainable farming approaches. 𝐒𝐤𝐢𝐥𝐥 𝐃𝐞𝐯𝐞𝐥𝐨𝐩𝐦𝐞𝐧𝐭: Collaborate with a diverse team, acquiring exceptional experience in applying AI to address real-world challenges. 𝐌𝐚𝐤𝐞 𝐚 𝐌𝐞𝐚𝐧𝐢𝐧𝐠𝐟𝐮𝐥 𝐃𝐢𝐟𝐟𝐞𝐫𝐞𝐧𝐜𝐞: Play a vital role in upgrading Nepal's agricultural data infrastructure, directly supporting the lives of those predominantly engaged in agriculture. 𝐖𝐡𝐨 𝐖𝐞 𝐍𝐞𝐞𝐝? We are in search of data scientists and AI engineers skilled in optical character recognition technology and language translation. Experience with the Nepali language and Devanagari script is a plus, though not required. 𝐃𝐚𝐭𝐚 𝐀𝐭 𝐘𝐨𝐮𝐫 𝐃𝐢𝐬𝐩𝐨𝐬𝐚𝐥 Access to: - Templates of loan documents. - Over 300 financial loan documents, both filled and unfilled. - Detailed farmer diaries and livestock documents outlining agricultural activities. 𝐋𝐚𝐮𝐧𝐜𝐡 𝐃𝐞𝐭𝐚𝐢𝐥𝐬 The challenge begins on July 2nd, with application submissions closing on June 30th. For more information and to apply, visit: https://lnkd.in/eiBe8hKj Not an expert? Join our mentorship track to test your skills and learn to apply AI in practice. Secure your spot today! 😄 https://lnkd.in/eiBe8hKj #AIForGood #AgricultureTechnology #DataScience #AIChallenge

🚀 Exciting Update on the AI for Women Farmers Challenge! 🌾 Last week, we successfully concluded the first phase of our initiative, focused on developing advanced AI tools to support women farmers. Here’s a brief overview of our achievements: 1️⃣ 🔍 Text Detection Model: Our model skillfully identifies and categorizes text within documents as either printed or handwritten, efficiently routing it to our OCR specialists for further processing. 2️⃣ 📖 Devenagari OCR Model: Employing a sophisticated sliding window technique, this model meticulously classifies each letter, assembling them into words ready for translation. 3️⃣ 🌍 Translation & NER Pipeline: This system is designed to not only translate text but also to extract meaningful information, enhancing our understanding of the document contents. In the second AI for Women Farmers Challenge, we're excited to enhance these models further and seamlessly integrate them into a comprehensive pipeline. 🤝 Join The Challenge: We’re inviting passionate individuals to contribute to this Challenge. Help us empower the agricultural cooperatives of Nepal by signing up here: https://lnkd.in/eiBe8hKj The next Challenge will kick off on the 2nd of July and will take 10 weeks to complete. This is a fantastic opportunity to apply AI in practical settings and expand your skills in the field! #AIforGood #OCR #NepaliFarmers #Agriculture

Check out the results from our AI against Oil Spills Challenge! ⛴️ This detailed case study explores how AI models like SAM and Mask-RCNN can be implemented to help oil spill clean-up by improving the speed and accuracy of volume estimations through segmentation masks. In collaboration with Rijkswaterstaat, we've harnessed drone-captured imagery to train computer vision models that can significantly enhance oil spill assessments, allowing for faster and more effective clean-up operations. Despite challenges like poor visibility and inconsistent annotations, our team managed to refine the dataset, leading to more accurate models and, potentially, better environmental protection. 👉 Read the full story and check out other Challenges on our platform: https://lnkd.in/gBQ6zhZt #AI #EnvironmentalProtection #Innovation #OilSpillCleanup #AIForGood