Region-based convolutional networks for end-to-end detection of agricultural mushrooms
In: Mansouri A., El Moataz A., Nouboud F., Mammass D. (eds) Image and Signal Processing. ICISP 2018. Lecture Notes in Computer Science, vol 10884. Springer, Cham. pp. 319-328
The article is available here at a cost.
Abstract: Olpin, A.J., Dara R., Stacey D. and M. Kashkoush. Conventional image processing techniques have been applied to the field of agricultural machine vision for the purposes of identifying crops for quality control, weed detection, automated spraying and harvesting. With the recent advancements in computational hardware Region-based Convolutional Networks have met with varying levels of success in the area of object detection and classification. In this study we found that a Region-based Convolutional Neural Network was able to achieve a 92% accuracy rating while a Region-based Fully Convolutional Network was able to achieve an 87% accuracy rating in the area of object detection operating on a newly create agricultural mushroom dataset.