Cromai unveils app for enhanced weed monitoring in Brazil

The Brazilian startup Cromai has introduced a new application designed to revolutionize weed monitoring in agriculture. Named Scan Weed Cana, the app is engineered to detect, geolocate, classify, and identify various weed types using advanced artificial intelligence, enabling more targeted weed management practices.

Unveiled at the 10th ConBAP (Brazilian Congress of Precision and Digital Agriculture), the app facilitates direct navigation to weed infestation points within agricultural fields. Cromai’s technology utilizes a comprehensive database of tens of millions of images to automatically pinpoint and categorize up to six distinct types of invasive plants. These include climbers, castor beans, low and tall grasses, other broadleaf plants, and a general category for undefined weeds, using drone-captured images processed through its Atlas platform.

Henrique Del Papa, COO of Cromai, highlighted the app’s precision, noting that it allows users to discern the exact location and severity of infestations, thereby enabling them to make more efficient and cost-effective decisions regarding field management. The application also features offline GPS capabilities, streamlining the process of locating infestation spots and functioning as a digital organizer for field data.

The Atlas platform additionally introduces the Activity Calendar, an interactive tool that facilitates comprehensive management planning. “It allows for scheduling all mapping and application activities directly on your operational calendar and monitoring the progress of field applications in a fully integrated environment,” explained Del Papa.

Cromai’s technology suite extends beyond weed detection, offering solutions like Cromai Sentinel for detecting vegetable impurities in sugarcane loads and Cromai Scan Growline for identifying planting failures in sugarcane fields using drone imagery and specialized algorithms. Cromai’s technology has been shown to decrease herbicide use by an average of 56% and water usage by 95%, while also minimizing tractor time in the field and reducing fuel consumption and carbon dioxide emissions.