scholarly journals Firearm Classification using Neural Networks on Ring of Firing Pin Impression Images

2013 ◽  
Vol 1 (3) ◽  
pp. 27-34 ◽  
Author(s):  
SAADI Bin Ahmad Kamaruddin ◽  
Nor AZURA MD Ghanib ◽  
Choong-Yeun Liong ◽  
Abdul AZIZ Jemain
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Transfer Functions ◽  
Training Algorithm ◽  
Backpropagation Neural Network ◽  
Training Set ◽  
Classification Result ◽  
Image Region ◽  
Geometric Moments ◽  
Validation Set

This paper implements two layer neural networks with different feedforward backpropagation algorithms for better performance of firearm classification us-ing numerical features from the ring image. A total of 747 ring images which are extracted from centre of the firing pin impression have been captured from five different pistols of the Parabellum Vector SPI 9mm model. Then, based on finding from the previous studies, the six best geometric moments numerical fea-tures were extracted from those ring images. The elements of the dataset were further randomly divided into the training set (523 elements), testing set (112 el-ements) and validation set (112 elements) in accordance with the requirement of the supervised learning nature of the backpropagation neural network (BPNN). Empirical results show that a two layer BPNN with a 6-7-5 configura-tion and tansig/tansig transfer functions with ‘trainscg’ training algorithm has produced the best classification result of 98%. The classification result is an improvement compared to the previous studies as well as confirming that the ring image region contains useful information for firearm classification.

scholarly journals Convolutional Neural Networks and Geometric Moments to Identify the Bilateral Symmetric Midplane in Facial Skeletons from CT Scans

Biology ◽  
2021 ◽  
Vol 10 (3) ◽  
pp. 182
Author(s):  
Rodrigo Dalvit Carvalho da Silva ◽  
Thomas Richard Jenkyn ◽  
Victor Alexander Carranza
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Surgical Planning ◽  
Good Accuracy ◽  
Bilateral Symmetry ◽  
Ct Scans ◽  
Facial Skeleton ◽  
Intact Side ◽  
Geometric Moments ◽  
The Face

In reconstructive craniofacial surgery, the bilateral symmetry of the midplane of the facial skeleton plays an important role in surgical planning. Surgeons can take advantage of the intact side of the face as a template for the malformed side by accurately locating the midplane to assist in the preparation of the surgical procedure. However, despite its importance, the location of the midline is still a subjective procedure. The aim of this study was to present a 3D technique using a convolutional neural network and geometric moments to automatically calculate the craniofacial midline symmetry of the facial skeleton from CT scans. To perform this task, a total of 195 skull images were assessed to validate the proposed technique. In the symmetry planes, the technique was found to be reliable and provided good accuracy. However, further investigations to improve the results of asymmetric images may be carried out.

Copy-Move Forgery Localization Using Convolutional Neural Networks and CFA Features

2018 ◽  
Vol 10 (4) ◽  
pp. 140-155 ◽  
Author(s):  
Lu Liu ◽  
Yao Zhao ◽  
Rongrong Ni ◽  
Qi Tian
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Image Processing ◽  
Color Filter ◽  
Interpolation Algorithm ◽  
Post Processing ◽  
Classification Result ◽  
Filter Array ◽  
Cfa Interpolation ◽  
Learned Features

This article describes how images could be forged using different techniques, and the most common forgery is copy-move forgery, in which a part of an image is duplicated and placed elsewhere in the same image. This article describes a convolutional neural network (CNN)-based method to accurately localize the tampered regions, which combines color filter array (CFA) features. The CFA interpolation algorithm introduces the correlation and consistency among the pixels, which can be easily destroyed by most image processing operations. The proposed CNN method can effectively distinguish the traces caused by copy-move forgeries and some post-processing operations. Additionally, it can utilize the classification result to guide the feature extraction, which can enhance the robustness of the learned features. This article, per the authors, tests the proposed method in several experiments. The results demonstrate the efficiency of the method on different forgeries and quantifies its robustness and sensitivity.

scholarly journals Evaluation of Power Insulator Detection Efficiency with the Use of Limited Training Dataset

2020 ◽  
Vol 10 (6) ◽  
pp. 2104
Author(s):  
Michał Tomaszewski ◽  
Paweł Michalski ◽  
Jakub Osuchowski
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Object Detection ◽  
Convolutional Neural Network ◽  
Deep Neural Networks ◽  
Detection Efficiency ◽  
Training Data ◽  
Training Dataset ◽  
Training Set ◽  
Convolutional Network

This article presents an analysis of the effectiveness of object detection in digital images with the application of a limited quantity of input. The possibility of using a limited set of learning data was achieved by developing a detailed scenario of the task, which strictly defined the conditions of detector operation in the considered case of a convolutional neural network. The described solution utilizes known architectures of deep neural networks in the process of learning and object detection. The article presents comparisons of results from detecting the most popular deep neural networks while maintaining a limited training set composed of a specific number of selected images from diagnostic video. The analyzed input material was recorded during an inspection flight conducted along high-voltage lines. The object detector was built for a power insulator. The main contribution of the presented papier is the evidence that a limited training set (in our case, just 60 training frames) could be used for object detection, assuming an outdoor scenario with low variability of environmental conditions. The decision of which network will generate the best result for such a limited training set is not a trivial task. Conducted research suggests that the deep neural networks will achieve different levels of effectiveness depending on the amount of training data. The most beneficial results were obtained for two convolutional neural networks: the faster region-convolutional neural network (faster R-CNN) and the region-based fully convolutional network (R-FCN). Faster R-CNN reached the highest AP (average precision) at a level of 0.8 for 60 frames. The R-FCN model gained a worse AP result; however, it can be noted that the relationship between the number of input samples and the obtained results has a significantly lower influence than in the case of other CNN models, which, in the authors’ assessment, is a desired feature in the case of a limited training set.

scholarly journals Machine-learning in astronomy

2014 ◽  
Vol 10 (S306) ◽  
pp. 279-287 ◽  
Author(s):  
Michael Hobson ◽  
Philip Graff ◽  
Farhan Feroz ◽  
Anthony Lasenby
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Neural Networks ◽  
Gamma Ray ◽  
Neural Network Training ◽  
Training Algorithm ◽  
Data Description ◽  
Astronomical Data ◽  
Machine Learning Methods ◽  
Network Training

AbstractMachine-learning methods may be used to perform many tasks required in the analysis of astronomical data, including: data description and interpretation, pattern recognition, prediction, classification, compression, inference and many more. An intuitive and well-established approach to machine learning is the use of artificial neural networks (NNs), which consist of a group of interconnected nodes, each of which processes information that it receives and then passes this product on to other nodes via weighted connections. In particular, I discuss the first public release of the generic neural network training algorithm, calledSkyNet, and demonstrate its application to astronomical problems focusing on its use in the BAMBI package for accelerated Bayesian inference in cosmology, and the identification of gamma-ray bursters. TheSkyNetand BAMBI packages, which are fully parallelised using MPI, are available athttp://www.mrao.cam.ac.uk/software/.

scholarly journals Prediksi Pemakaian Listrik Kelompok Tarif Menggunakan Jaringan Syaraf Tiruan dan ARIMA

2013 ◽  
Vol 7 (2) ◽  
pp. 189
Author(s):  
Silviani E Rumagit ◽  
Azhari SN
Keyword(s):  
Neural Network ◽  
Artificial Intelligence ◽  
Neural Networks ◽  
Statistical Method ◽  
Moving Average ◽  
Electric Energy ◽  
Backpropagation Neural Network ◽  
The Difference ◽  
The Government ◽  
Increasing Demand

AbstrakLatar Belakang penelitian ini dibuat dimana semakin meningkatnya kebutuhan listrik di setiap kelompok tarif. Yang dimaksud dengan kelompok tarif dalam penelitian ini adalah kelompok tarif sosial, kelompok tarif rumah tangga, kelompok tarif bisnis, kelompok tarif industri dan kelompok tarif pemerintah. Prediksi merupakan kebutuhan penting bagi penyedia tenaga listrik dalam mengambil keputusan berkaitan dengan ketersediaan energi listik. Dalam melakukan prediksi dapat dilakukan dengan metode statistik maupun kecerdasan buatan.            ARIMA merupakan salah satu metode statistik yang banyak digunakan untuk prediksi dimana ARIMA mengikuti model autoregressive (AR) moving average (MA). Syarat dari ARIMA adalah data harus stasioner, data yang tidak stasioner harus distasionerkan dengan differencing. Selain metode statistik, prediksi juga dapat dilakukan dengan teknik kecerdasan buatan, dimana dalam penelitian ini jaringan syaraf tiruan backpropagation dipilih untuk melakukan prediksi. Dari hasil pengujian yang dilakukan selisih MSE ARIMA, JST dan penggabungan ARIMA, jaringan syaraf tiruan tidak berbeda secara signifikan. Kata Kunci— ARIMA, jaringan syaraf tiruan, kelompok tarif.  AbstractBackground this research was made where the increasing demand for electricity in each group. The meaning this group is social, the household, business, industry groups and the government fare. Prediction is an important requirement for electricity providers in making decisions related to the availability of electric energy. In doing predictions can be made by statistical methods and artificial intelligence.            ARIMA is a statistical method that is widely used to predict where the ARIMA modeled autoregressive (AR) moving average (MA). Terms of ARIMA is the data must be stationary, the data is not stationary should be stationary  use differencing. In addition to the statistical method, predictions can also be done by artificial intelligence techniques, which in this study selected Backpropagation neural network to predict. From the results of tests made the difference in MSE ARIMA, ANN and merging ARIMA, artificial neural networks are not significantly different. Keyword—ARIMA, neural network, tarif groups

scholarly journals Modeling Nonlinear Transfer Functions from Speech Envelopes to Encephalography with Neural Networks

2019 ◽  
Vol 11 (4) ◽  
pp. 1 ◽  
Author(s):  
Tobias de Taillez ◽  
Florian Denk ◽  
Bojana Mirkovic ◽  
Birger Kollmeier ◽  
Bernd T. Meyer
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Artificial Neural Network ◽  
Linear Models ◽  
Transfer Functions ◽  
Event Related Potentials ◽  
Stimulus Presentation ◽  
Related Potentials ◽  
Linear And Nonlinear ◽  
Artificial Neural

Diferent linear models have been proposed to establish a link between an auditory stimulus and the neurophysiological response obtained through electroencephalography (EEG). We investigate if non-linear mappings can be modeled with deep neural networks trained on continuous speech envelopes and EEG data obtained in an auditory attention two-speaker scenario. An artificial neural network was trained to predict the EEG response related to the attended and unattended speech envelopes. After training, the properties of the DNN-based model are analyzed by measuring the transfer function between input envelopes and predicted EEG signals by using click-like stimuli and frequency sweeps as input patterns. Using sweep responses allows to separate the linear and nonlinear response components also with respect to attention. The responses from the model trained on normal speech resemble event-related potentials despite the fact that the DNN was not trained to reproduce such patterns. These responses are modulated by attention, since we obtain significantly lower amplitudes at latencies of 110 ms, 170 ms and 300 ms after stimulus presentation for unattended processing in contrast to the attended. The comparison of linear and nonlinear components indicates that the largest contribution arises from linear processing (75%), while the remaining 25% are attributed to nonlinear processes in the model. Further, a spectral analysis showed a stronger 5 Hz component in modeled EEG for attended in contrast to unattended predictions. The results indicate that the artificial neural network produces responses consistent with recent findings and presents a new approach for quantifying the model properties.

scholarly journals A Green Prospective for Learned Post-processing in Sparse-view Tomographic Reconstruction

2021 ◽  
Author(s):  
Elena Morotti ◽  
Davide Evangelista ◽  
Elena Loli Piccolomini
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Deep Learning ◽  
Tomographic Reconstruction ◽  
Research Field ◽  
Filtered Backprojection ◽  
Post Processing ◽  
Training Set ◽  
Imaging Reconstruction ◽  
Active Research

Deep Learning is developing interesting tools which are of great interest for inverse imaging applications. In this work, we consider a medical imaging reconstruction task from subsampled measurements, which is an active research field where Convolutional Neural Networks have already revealed their great potential. However, the commonly used architectures are very deep and, hence, prone to overfitting and unfeasible for clinical usages. Inspired by the ideas of the green-AI literature, we here propose a shallow neural network to perform an efficient Learned Post-Processing on images roughly reconstructed by the filtered backprojection algorithm. The results obtained on images from the training set and on unseen images, using both the non-expensive network and the widely used very deep ResUNet show that the proposed network computes images of comparable or higher quality in about one fourth of time.

UNSUPERVISED COMPETITIVE LEARNING IN NEURAL NETWORKS

1989 ◽  
Vol 01 (02) ◽  
pp. 177-186
Author(s):  
Atilla E. Gunhan ◽  
László P. Csernai ◽  
Jørgen Randrup
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Unsupervised Learning ◽  
Lateral Inhibition ◽  
Competitive Learning ◽  
Learning Ability ◽  
Neural Systems ◽  
Training Set

We study an idealized neural network that may approximate certain neurophysiological features of natural neural systems. The network contains a mutual lateral inhibition and is subjected to unsupervised learning by means of a Hebb-type learning principle. Its learning ability is analysed as a function of the strength of lateral inhibition and the training set.

scholarly journals Forward-Backward Convolutional Recurrent Neural Networks and Tag-Conditioned Convolutional Neural Networks for \\Weakly Labeled Semi-supervised Sound Event Detection

2021 ◽  
Author(s):  
Janek Ebbers ◽  
Reinhold Haeb-Umbach
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Recurrent Neural Network ◽  
Event Detection ◽  
Forward Direction ◽  
Time Step ◽  
Sound Event ◽  
Sound Event Detection ◽  
Validation Set ◽  
Classi Fication

In this paper we present our system for thedetection and classi-fication of acoustic scenes and events (DCASE) 2020 ChallengeTask 4: Sound event detection and separation in domestic envi-ronments. We introduce two new models: the forward-backwardconvolutional recurrent neural network (FBCRNN) and the tag-conditioned convolutional neural network (CNN). The FBCRNNemploys two recurrent neural network (RNN) classifiers sharing thesame CNN for preprocessing. With one RNN processing a record-ing in forward direction and the other in backward direction, thetwo networks are trained to jointly predict audio tags, i.e., weak la-bels, at each time step within a recording, given that at each timestep they have jointly processed the whole recording. The pro-posed training encourages the classifiers to tag events as soon aspossible. Therefore, after training, the networks can be appliedto shorter audio segments of, e.g.,200 ms, allowing sound eventdetection (SED). Further, we propose a tag-conditioned CNN tocomplement SED. It is trained to predict strong labels while using(predicted) tags, i.e., weak labels, as additional input. For train-ing pseudo strong labels from a FBCRNN ensemble are used. Thepresented system scored the fourth and third place in the systemsand teams rankings, respectively. Subsequent improvements allowour system to even outperform the challenge baseline and winnersystems in average by, respectively,18.0 %and2.2 %event-basedF1-score on the validation set. Source code is publicly available athttps://github.com/fgnt/pb_sed

Sign in / Sign up

Export Citation Format

Share Document