Every year, the agriculture sector loses billions of dollars in products and assets to both wildfires and building fires. Fire destroys croplands, buildings, equipment, livestock, silage and other bulk stores and even kills farm workers.
Lemay.ai has a proven track record of successfully creating image recognition systems that can use satellite images, camera images, video, and/or industrial internet of things (IIoT) sensors to detect smoke and fire. We also offer solutions that predict fire by augmenting visual data processing using meteorological, soil, geographic information systems (GIS) and other sources.
When you wrap it all up in an intuitive user interface that delivers alerts about changing risks and integrates with other platforms, you have a system that saves lives and businesses.
The presence of disease can destroy entire crops. In some cases, the only potential defence is to detect the problem early enough that it’s still manageable. But even if you could spend all day, every day inspecting crops, you still might not see those first warning signs of disease.
Our machine learning systems solve that problem. They can bring together satellite photos, images from static cameras, or drone images and videos. Once the images are collected, our deep learning image recognition models can be trained to detect any changes to plant health that can provide that essential early warning. Images and videos can factor in light levels, leaf position, and leaf orientation for startlingly accurate results.
Changes to the leaves in colour or texture, evidence of damage (like holes) or the appearance of actual insects can be captured in images and compared to extensive datasets to identify potential red flags for investigation and treatment. Datasets can be augmented to include any threats particular to your crop and locale.
Sensor fusion combines data from a variety of sources to provide more information and greater accuracy than from one source alone. In agricultural systems, we typically combine information from sources like visual-spectrum images, infrared images, ultraviolet images, water sensors, pH level indicators, electrical conductivity sensors for salinity monitoring, oxidation-reduction potential (ORP) sensors to monitor disinfectant levels, sensors for key nutrient levels, thermometers, and more.
When combined with image recognition systems, sensors can tell if plants are receiving the right amount of water and nutrients, and what adjustments you should make. For outdoor crops, adjustments can factor in information from weather forecasts and topographical information for fields.
Timing your harvest well can make a big difference in product quality and prices. Assessing the readiness of green fruit can be especially challenging, especially if you need to monitor different areas of fields with different microclimates, or different fields in different locations.
By combining image recognition, historic crop information, weather information, and other data sources Lemay.ai can create systems that help even the most experienced agriculturalists determine crop readiness.
Precision farming is the use of advanced technology—including some of the systems we’ve described above—to maximize yields as efficiently as possible. Part of the goal is to help smaller farmers stay in business, and the other goal is better sustainability from the more targeted use of resources like water, fertilizers, and pesticides.
Machine learning solutions have a big role to play in precision farming because they help solve two fundamental problems:
Large-scale information gathering
Preliminary problem identification
In other words, they do the legwork so that people have more time to confirm issues and to make decisions.
Lemay.ai culture is built on creating new solutions to organizational problems. Let us help you find a new way to go beyond the possible.
Our goal is to be as helpful as possible.