SORTASI BIJI KOPI DENGAN IMAGE PROCESSING MENGGUNAKAN CONVOLUTIONAL NEURAL NETWORK

Image processing is a science in the field of komputer vision which is used to produce a system that uses human visual language in general. Image processing is a method or technique that can be used to process an image data that is filled in to obtain certain information about the object observed in image processing, the existing image is processed so that the image becomes easier to process while komputer vision has the main objective, namely to make a useful decision about the real physical object obtained from the device or sensor. Current image processing applications have been widely applied in other world agriculture for physical damage to fungal contamination of maize, classifying apples according to size, to see the rate of bruising on salak, for grading fruits and vegetables. one of the other agricultural fields is the quality of the coffee beans. In improving the quality of coffee beans in Indonesia, they still use the manual method so that it is less efficient and effective and has non-uniform results This study uses robusta coffee beans with 100 images of data. Then the image processing is carried out using roboflow so that the data becomes 605 images, the next process is to build a network architecture model using a convolutional neural network using the python programming language to carry out the sorting process using a servo motor The results of the research that have been carried out show that sorting coffee beans using a convolutional neural network has been successfully carried out with the results of training accuracy of 99% and validation accuracy of 93% and the average accuracy of tool success is 79%.