Learn more: Simplify Model Development with Hosted Training
Leverage model training infrastructure for your custom computer vision tasks and get into production quickly.
For the food processing industry, enhancing yield and reducing waste is a key business challenge. FloVision leverages AI-powered insights to help food processors reduce waste, boost efficiency, and analyze critical production data. By integrating solutions directly into processing lines, FloVision helps organizations evaluate products and processes in real time.
"Our mission is to create a financially and environmentally sustainable future for the food industry," stated Rian McDonnell, founder and CEO of FloVision. "To achieve that goal, we’re equipping food processors with AI-assisted automation and analytics that aim to streamline production and increase yield from start to finish."
FloVision uses machine learning, specifically computer vision, to analyze imagery from food processing lines. This analysis enables precise yield measurement against targets, automated quality grading and sorting, financial performance tracking, and better traceability.
A key aspect is the creation of real-time feedback loops, providing timely information to operators and supervisors for immediate corrective actions. This ranges from real-time quality alerts to guided processing systems. "We help people all along the production line extract essential information about operations," explained McDonnell. "This enables them to identify and address issues or improve processes, such as catching quality issues earlier in the process or better utilizing raw materials."
One example of the impact of FloVision’s platform comes from a leading beef processor. By using a mixture of guided systems and visual analysis, the organization reduced the amount of protein lost to overtrimming. The corrected trimming process resulted in a significant improvement in yield and profitability. In addition, they observed a 6.7% increase in trimming speed allowing for increased throughput.
By enhancing yield and efficiency in food processing lines, FloVision also helps clients pursue their sustainability goals. “We’ve seen situations where, by reducing waste, a single food processor can eliminate the equivalent of millions of tons of carbon per year. On a personal level, it feels great seeing AI-powered insights not only improve profitability for organizations, but also contribute to their sustainability goals,” McDonnell explained.
A critical component of FloVision's ability to deliver these impactful solutions is their strategic use of vision AI. "Computer vision is a fundamental component of our offering," stated McDonnell. "We’ve created vision models capable of identifying nuanced factors in a range of food items. Our vision models not only detect defects and measure dimensions, but can also evaluate slight variations in color and texture. This visual intelligence is at the heart of our analytics and automation solutions."
Initially, FloVision managed computer vision models with a collection of disparate tools and cloud services, but later migrated to an end-to-end solution with Roboflow. Now the team can annotate images, train purpose-built models, integrate foundation models, and get everything into production using one platform.
“Running our vision models through different notebooks and clouds resulted in added overhead and costs, especially when maintaining and updating our core offering,” McDonnell said. “Since moving the majority of our vision models to a centralized platform like Roboflow, we’ve increased the cadence in rolling out new features and tools for our clients. Do we need a new model to evaluate this kind of food? We spin up the platform, train overnight, and the next morning it works.”
To better assist food processors across industries, FloVision offers the ability to run AI models in the cloud or edge devices depending on their latency, bandwidth, and usage requirements. “In some processing environments, it’s important to detect issues in real time, without waiting for imagery to be transferred over the internet,” McDonnell explained. “We started off using Roboflow for the annotation tools, but were pleasantly surprised when we discovered it supported running models in the cloud as well as on-premises devices.”
Having helped analyze over 20 million pounds of food, the team at FloVision is committed to bringing further innovations to the food analytics space. "Seeing our work result in improved yields and less waste from the industry is gratifying on many levels,” McDonnell said. “By bringing vision AI to food processors, FloVision is not only helping them improve their bottom line but also contributing to a more sustainable and efficient global food system.”
FloVision Solutions is a food analytics platform optimizing the food chain. By installing advanced camera and sensor systems in food production facilities, FloVision uses AI to measure critical parameters such as yield, waste, and quality control, providing actionable insights to enhance efficiency and sustainability.
