Perkembangan teknologi saat ini memacu penerapan pengenalan pola dalam berbagai bidang, seperti pengenalan pola tanda tangan, sidik jari, wajah, dan tulisan tangan. Tulisan tangan manusia memiliki perbedaan antara satu dengan yang lainnya dan sering terjadi tulisan tangan susah terbaca atau susah dikenali dan hal ini dapat menghambat aktivitas sehari-hari, misalnya aktivitas transaksi yang memerlukan tulisan tangan. Bahkan salah satu ciri biometrik pada setiap orang adalah tulisan tangan. Salah satu metode yang dapat digunakan untuk mengenali pola tulisan tangan dalam bidang ilmu komputer adalah jaringan syaraf tiruan (JST) dengan algoritma pembelajaran adalah backpropagation. Jaringan syaraf tiruan mampu mengenali sesuatu dengan berbasis masa lalu. Artinya data masa lalu akan dipelajari sehingga mampu memberi keputusan terhadap data baru. Untuk mengenali pola tulisan tangan menggunakan jaringan syaraf tiruan, maka ciri-ciri dari objek tulisan tangan diekstrak menggunakan moment invariant. Hasil pelatihan menggunakan jaringan syaraf tiruan menunjukkan bahwa nilai koefisien korelasi didapatkan pada jumlah neuron hidden layer sebesar 30. Nilai koefisien korelasi tertinggi sebesar 0,61382. Hasil pengujian pada data uji didapatkan tingkat akurasi sebesar 11,67% dari total data uji. Kata kunci: pola tulisan tangan, jaringan syaraf tiruan, angka, moment invariant
Plagiarism is the use of data, language and writing without including the original author or source. The place where palgiate practice occurs most often is the academic environment. In the academic world, the most frequently plagiarized thing is scientific work, for example thesis. To minimize the practice of plagiarism, it is not enough to just remind students. Therefore we need a system or application that can help in measuring the level of similarity of student thesis proposals in order to minimize plagiarism practice. In computer science, the Rabin-Karp algorithm can be used in measuring the level of similarity of texts. The Rabin-Karp algorithm is a string matching algorithm that uses a hash function as a comparison between the search string (m) and substrings in text (n). The Rabin-Karp algorithm is a string search algorithm that can work for large data sizes. The test results show that the use of values on k-gram has an effect on the results of the measurement of similarity levels. In addition, it was also found that the use of the value 5 on k-gram was faster in executing than the values 4 and 6.
Implementation of computer vision can be done in the introduction of images or pictures of characters of numbers or letters. Based on this, then the computer vision can be used in the introduction of numbers on the electric meter or commonly called kWh meter. The underlying thing for the electric meter to be the object of research is to look at the situation, where the electric meter recorder keeps the record using the camera. Furthermore, the value shown on the electric meter will be inputted manually. Manual input requires a relatively long time because the amount of electricity meter input value is not small data. One method that can be used in recognizing the shape of the image in computer vision is the invariant moment. The results of this study indicate that the quality of the image gives effect, both in terms of the extraction of features and the accuracy of the recognition of the figure on the image of the electric meter. In addition to this, the threshold value of the euclidian distance method should also be used to limit the recognition process.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.