Detection of Corn Maturity Based on Color Composition Using the K-Nearest Neighbor Algorithm
AbstractCorn is one of side dish in Indonesia. The utilization is also diverse and different at each level of maturity. In agriculture, the determination of maize maturity is usually done manually by visual observation and aspects of other agricultural sciences that have been mastered by farmers and from the experience they have gained. Manually observing can be different for each farmer, which will cause the products produced to be inconsistent due to visual limitations, fatigue and different perceptions of each observer. Corn maturity level consists of 4 levels and each level has different savings. For example, young corn is commonly used for vegetables, rather old corn is made roasted corn and dry corn for chicken feed. Because of these shortcomings, an application was made to classify the maturity of corn to get results more objectively. Farmers can find out the maturity level of corn from the image of corn taken by an Android smartphone. On the website, only the administrator has access to see and add training data. Input from the image is taken using an Android camera which is expected to be able to distinguish the maturity of corn. So from the process of testing accuracy 510 data training with the nearest neighbor value range of k 5-15 using k-Fold Cross Validation, the highest level of accuracy is 90.98% for the value of neighbor is k = 13.
Copyright info for authors
1. Authors hold the copyright in any process, procedure, or article described in the work and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
2. Authors retain publishing rights to re-use all or portion of the work in different work but can not granting third-party requests for reprinting and republishing the work.
3. Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) as it can lead to productive exchanges, as well as earlier and greater citation of published work.