Click on “Download PDF” for the PDF version or on the title for the HTML version.

If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

Abstract. Strawberry yield prediction is important for growers because accurate yield prediction can help them to hire the proper amount of labor for strawberry harvesting. Strawberry flower is one of the most important indicators for strawberry yield. However, the traditional manual counting method is time-consuming and impractical for a large strawberry field. Therefore, a fast and accurate strawberry flower counting method is needed for strawberry growers. In this study, a flower detection system was developed using fluorescence imaging under a controlled lighting environment. Four cameras and four UVA lamps (315 nm - 380 nm) were mounted on a ground imaging cart, and a lightproof and waterproof canvas was used to cover the ground imaging cart to keep the illumination uniform. When the cart was moving in the field, the cameras collected strawberry flower images. Because of the fluorescence characteristic of strawberry flower, the flower looked brighter than other objects such as strawberry fruit, leaf, plant stem, soil, etc., which made strawberry flower detection much easier than the images collected under natural lighting conditions. Based on the preliminary results, basic image processing methods (threshold segmentation, morphological operations, and object size analysis) were able to detect most of the flowers with an accuracy of 90%. However, many non-flower objects were classified as flowers, which would require further investigation in the near future.