Fingerprint Pattern Classification Using Convolution Neural Network

Jeon, Wang-Su; Rhee, Sang-Yong

doi:10.5391/ijfis.2017.17.3.170

Cited by 43 publications

(23 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…This training mechanism for convolution filters has made significant performance improvements in pattern recognition tasks such as image classification, character recognition, semantic segmentation, engineering classification, and so on. [2][3][4] RNNs are models that use both their current state and current input to determine its next state. 5 Traditional RNNs suffer from both vanishing gradient problem and exploding gradient problem.…”

Section: Deep Neural Networkmentioning

confidence: 99%

“…To help model development, some graphical user interface (GUI)-based tools have been developed with which developers can construct a deep neural network model in a drag-and-drop manner. 4 It is convenient to use such tools when the number of layers is small enough to be drawn on the display window. However, they cannot handle large models on a small display window; hence, they are only appropriate for designing deep learning models with a relatively small number of layers.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Deep neural network model construction with interactive code reuse and automatic code transformation

Lee

Park

Hwang

et al. 2019

Concurrency and Computation

View full text Add to dashboard Cite

Summary Many application developers have been recently interested in applying deep learning techniques to their works but have little knowledge and experience on them. This paper presents the methods for a GUI‐based modeling tool to easily build deep learning models and automatically transform them into executable program codes. For reuse of existing deep learning codes, it introduces a method that imports such a program code, extracts the deep learning model architecture, and transforms it into a graphical representation that can be modified on a graphical interface. Meanwhile, a deep learning model with many layers is difficult to be visualized on a small display screen. To handle the difficulty, it proposes a method to identify the submodules called articulable subgraphs in a deep learning model and to organize the deep learning model into a hierarchical architecture using the nesting relationships of articulable subgraphs. It introduces a method to identify frequent substractures as well as articulable subgraphs as building blocks for deep learning models for levelwise views. The hierarchical representation easily enables to build deep learning models with many layers. The GUI‐based modeling tool employing the proposed methods makes nonexpert developers easily use deep learning techniques in their practical applications.

show abstract

Section: Deep Neural Networkmentioning

confidence: 99%

mentioning

confidence: 99%

Deep neural network model construction with interactive code reuse and automatic code transformation

Lee

Park

Hwang

et al. 2019

Concurrency and Computation

View full text Add to dashboard Cite

show abstract

“…There are many application of CNN like image recognition, object detection, face recognition, fingerprint pattern classification etc. [12]. A standard CNN architecture consists of a combination of convolution layers (feed-forward layers) and pooling layers and after the last pooling layer, the network is connected to a fully connected layer.…”

Section: Introductionmentioning

confidence: 99%

Performance Analysis of Classification Techniques of Human Brain MRI Images

Chauhan

Choi

2019

IJFIS

View full text Add to dashboard Cite

Artificial intelligence enhances the boundaries and capabilities of medical imaging. Hence, researchers are continuously attempting to develop an efficient and automated diagnosis system to increase the accuracy and performance to diagnose the brain abnormality. Therefore, it is required that a suitable method to diagnose and classify brain-related diseases such as Alzheimer disease, cancer, dementia, etc. Magnetic resonance imaging (MRI) is a powerful imaging technique in neuroscience for studying brain images. In the past years, many brain MRI classification techniques were proposed. Machine learning and deep learning have demonstrated a wonderful performance in the classification task. In this paper, the study of various brain MRI classification techniques has been provided. The aim of this study is to help the doctors/neurologists in selection of appropriate classification method based on several parameters like accuracy, computer complexity, and low training data availability. We also analyze and compare the performance of different classification methods based on several evaluation metrics.

show abstract

“…Shrein [3] used Convolutional Neural Networks (CNNs) model for the fingerprint image classification and showed that effective image preprocessing can greatly reduce the dimensionality of the problem. Jeon and Rhee [4] also proposed the use of a CNN model combined with an ensemble model and a batch normalization technique after improving the quality of fingerprint images. Perrier [5] reported that Google announced the release of Tensorflow.js, a Javascript implementation of Tensorflow.…”

Section: Introductionmentioning

confidence: 99%

An Online Integrated Fingerprint Image System

Huang¹,

Liu²,

Chen³

2019

IJMLC

View full text Add to dashboard Cite

Fingerprint analysis is one of the most commonly-used biometrics technologies, primarily used in forensic science to identify people. Due to various reasons, most fingerprint related studies are conducted in an isolated environment which makes it hard for researchers to collaborate across institutes. Web-based client/server applications have become an important component for international collaboration because such an application allows users from different locations to execute the same program on the same data. Such systems can be quickly developed and are easy to use. In this paper, a 3-tier integrated online fingerprint system was developed to manage and process fingerprint images in real-time. This system allows users all over the world to access the public NIST-14 fingerprint dataset and collaborate with other researchers to conduct the research task. The front-end of the system is an advanced web-based interface that allows users to view fingerprint images in different resolution, singular points, enter, and update the location of singular points. This system can also preserve the entered information in the back-end database.

show abstract

Fingerprint Pattern Classification Using Convolution Neural Network

Cited by 43 publications

References 5 publications

Deep neural network model construction with interactive code reuse and automatic code transformation

Deep neural network model construction with interactive code reuse and automatic code transformation

Performance Analysis of Classification Techniques of Human Brain MRI Images

An Online Integrated Fingerprint Image System

Contact Info

Product

Resources

About