Desarrollo de un algoritmo para detectar el efecto Bloom en los arándanos

This report focused on the recognition of blueberries using the segmentation method and the post selection of the blueberries identified as good. The image change method from RGB to HSV was applied, which allowed a better discrimination of blueberries in relation to the background and other elements present in the image. Successful segmentation was achieved by creating masks based on the values of hue, saturation and value (HSV). Classification based solely on the absolute size of blueberries may not be enough to guarantee an accurate and consistent classification. With the Hough transformation method, it will be possible to better discriminate blueberries by approaching them in a circular way. However, an image filter was used with acceptable Bloom concentration ranges for greater accuracy. In conclusion, this report highlights the success in the recognition of blueberries through the segmentation method and the presentation of the blueberries selected as good. With an average effectiveness of 98%, favorable results can be obtained so that the process can be used to mention as valid for the objectives set.