IJCER Online Submission
IJCER Paper Submission Format
Publication Charges - IJCER

Frequency: 12 issues per year
ISSN: 2250–3005 (online version)
Download Certificates
Call for paper...

International Journal of Computational Engineering Research

(IJCER)

 

Articles

 

Research Article  open access
Melody Generation Using Recurrent Neural Network
Sarthak Somvanshi || Vinit Vaidya || Chinmay Vartak || Dnyaneshwar Kapse
India
International Journal of Computational Engineering Research, Volume 11 ~ Issue 04 (April – 2021)

Abstract

In recent years, neural networks have been used to generate symbolic melodies. However, the long- term structure in the melody has posed great difficulty for designing a good model. In this paper, we present a hierarchical recurrent neural network for melody generation, which consists of three Long-Short-Term- Memory (LSTM) subnetworks working in a coarse-to- fine manner along time. Specifically, the three subnetworks generate bar profiles, beat profiles and notes in turn, and the output of the high-level subnetworks are fed into the low-level subnetworks are fed into the low-level subnetworks, serving as guidance for generating the finer-time-scale melody components in low-level subnetworks. Two human behaviour experiments demonstrate the advantage of this structure over the single-layer LSTM which attempts to learn all hidden structures in melodies.

Keywords:Machine Learning, Melody, Melody Generation Using Recurrent Neural Network, Literature Survey.

Research Article  open access
A Comparative study of Machine Learning Algorithms Using RDD Based Regression and Classification Methods
A.VETTRISELVI || N. GNANAMBIGAI || P. DINADAYALAN || S.SUTHA
India
International Journal of Computational Engineering Research, Volume 11 ~ Issue 04 (April – 2021)

Abstract

Today we live in the world of big data, where we find it difficult to store and process the data by the traditional devices. This paper main aim is to help the researchers and professional who are already familiar with machine learning but not experienced with MLlib package for both Regression and classification methods using RDD-based in big data. This paper provides a list of scenario. Firstly, we go on Hadoop environment and have to integrate and deploy spark on yarn, secondly we explore how to allocate resources dynamically executors cores and memory which improves the performance of spark application on YARN, Thirdly, we have implemented RDDs API in spark on YARN, which we evaluate throughout the experiments. Finally, we compare and evaluate few machine learning algorithms in spark using RDD-based regression.......

Keywords:machine learning, RDD, spark, Hadoop, Hibench, Regression, Classification, MAE, RMSE.

Research Article  open access
Touchless Fingerprint Recognition System
Navya Sheregar || Vrutti Mistry || Pooja Palekar || Tejas Phutane || Prof. Sanjna Repal
India
International Journal of Computational Engineering Research, Volume 11 ~ Issue 04 (April – 2021)

Abstract

Fingerprint Recognition system are the most widely used biometric identification. Touchless Fingerprint Recognition system is a new option for the old conventional touch-based fingerprint recognition system in the way that it uses a digital camera to acquire the fingerprint image. This is comfortable, inexpensive and now fast enough for practical use. This paper helps to find a new ideal solution to the problem in view of maintenance, hygiene and latent fingerprints. In this paper a touchless fingerprint recognition system based on a novel fingerprint minutiae matching algorithm is presented. The system consists of mainly three stages- pre-processing, feature extraction and matching stage. The extraction and matching performances are totally dependent on the quality of fingerprint images. Better quality images lead to better extraction and matching performances

Keywords:Touchless fingerprint image; Fingerprint pre-processing; Ridge filtering; Minutiae Extraction; False Matching Rate; False Non-matching Rate.

Research Article  open access
Traffic Sign Detection and Recognition (TSDR)
Kartik Jain || Mayank Sinha || Eraa || Mayank Vats || Mr. Pramod Sethy
India
International Journal of Computational Engineering Research, Volume 11 ~ Issue 04 (April – 2021)

Abstract

The world is evolving every day, as the people are continuously working to make things simpler and simpler by automating them and one of such task is the Advance Driver Assistance System(ADAS). The application of ADAS is 'Traffic Sign Detection And Recognition" (TSDR). TSDR is the system in which traffic signs are automatically detected and recognized. It plays a crucial role for the one who is driving the vehicle. As the driver needs to stay focused on the road while driving, the drivers might miss some of the road signs which can be dangerous for the driver of the vehicle as well as for other drivers. The TSDR will reduce this risk by automatically detecting the road sign using Computer Vision and machine learning algorithms such as Convolutional Neural Network (CNN) as well as recognizing them. This whole process will reduce the human efforts and the machine will accurately detect the sign without any human error.

Keywords:Computer Vision, Image Processing, CNN, Tensorflow, Traffic Sign Detection, Traffic Sign Recognition, Advance Driver Assistance System