<![CDATA[The quality of the fruit is largely determined by the level of ripeness contained by the fruit itself. Until now, determining the level of fruit maturity is still done manually, as a result there are differences in perceptions in determining the level of fruit maturity. Therefore we need a system that is able to classify fruit maturity automatically. This research was conducted on 4 objects, namely apples, oranges, mangoes, and tomatoes. The training was conducted with split data with a presentation 70:20:10 based on 4 test scenarios, the data was converted to RGB to L * a * b first and some were not converted and were immediately trained using CNN VGG16 with the transfer learning method where fine tuning would be done on the layer block 5 and modification of the classification layer using the Multi-SVM classifier. The highest accurasi reach 92% at scenario 4 with 90 data per class.]]>
Copyrights © 2023
