Testing of crop detection
Summary
Testing of crop detection
S00039
Version: 1.7
- arable farming
- horticulture
- tree crops
- viticulture
- test design
- test setup
- test execution
- collection of test data
- collection of test data
- performance evaluation
- physical system (possibly including software)
- software
- Location: at user’s premises, in Austria
- Offered by: JR
Description
This digital test is designed to assess the capabilities of a model in recognizing various types of crops accurately. Typically, this system is provided in the form of a docker container or a similar package, which allows for convenient and standardized testing across different platforms. Depending on the specifics of the system and the model under evaluation, key metrics such as accuracy, Intersection over Union (IoU), and the F1 score are used to measure performance. These metrics provide a comprehensive understanding of the model's effectiveness in crop recognition tasks.
For the purpose of this test, the ground truth data is crucial. This data can either be specifically recorded and stored in a dedicated data service for this test or it can be sourced from an existing benchmark data set (S00042).