Search results for "Convolution"

Temperate Fish Detection and Classification: a Deep Learning based Approach

2021

A wide range of applications in marine ecology extensively uses underwater cameras. Still, to efficiently process the vast amount of data generated, we need to develop tools that can automatically detect and recognize species captured on film. Classifying fish species from videos and images in natural environments can be challenging because of noise and variation in illumination and the surrounding habitat. In this paper, we propose a two-step deep learning approach for the detection and classification of temperate fishes without pre-filtering. The first step is to detect each single fish in an image, independent of species and sex. For this purpose, we employ the You Only Look Once (YOLO) …

FeatherCNN: Fast Inference Computation with TensorGEMM on ARM Architectures

2020

Deep Learning is ubiquitous in a wide field of applications ranging from research to industry. In comparison to time-consuming iterative training of convolutional neural networks (CNNs), inference is a relatively lightweight operation making it amenable to execution on mobile devices. Nevertheless, lower latency and higher computation efficiency are crucial to allow for complex models and prolonged battery life. Addressing the aforementioned challenges, we propose FeatherCNN – a fast inference library for ARM CPUs – targeting the performance ceiling of mobile devices. FeatherCNN employs three key techniques: 1) A highly efficient TensorGEMM (generalized matrix multiplication) routine is app…

Hybrid Deep Shallow Network for Assessment of Depression Using Electroencephalogram Signals

2020

Depression is a mental health disorder characterised by persistently depressed mood or loss of interest in activities resulting impairment in daily life significantly. Electroencephalography (EEG) can assist with the accurate diagnosis of depression. In this paper, we present two different hybrid deep learning models for classification and assessment of patient suffering with depression. We have combined convolutional neural network with Gated recurrent units (RGUs), thus the proposed network is shallow and much smaller in size in comparison to its counter LSTM network. In addition to this, proposed approach is less sensitive to parameter settings. Extensive experiments on EEG dataset shows…

Multiple Fault Diagnosis of Electric Powertrains Under Variable Speeds Using Convolutional Neural Networks

2018

Electric powertrains are widely used in automotive and renewable energy industries. Reliable diagnosis for defects in the critical components such as bearings, gears and stator windings, is important to prevent failures and enhance the system reliability and power availability. Most of existing fault diagnosis methods are based on specific characteristic frequencies to single faults at constant speed operations. Once multiple faults occur in the system, such a method may not detect the faults effectively and may give false alarms. Furthermore, variable speed operations render a challenge of analysing nonstationary signals. In this work, a deep learning-based fault diagnosis method is propos…

2021

Classification approaches that allow to extract logical rules such as decision trees are often considered to be more interpretable than neural networks. Also, logical rules are comparatively easy to verify with any possible input. This is an important part in systems that aim to ensure correct operation of a given model. However, for high-dimensional input data such as images, the individual symbols, i.e. pixels, are not easily interpretable. Therefore, rule-based approaches are not typically used for this kind of high-dimensional data. We introduce the concept of first-order convolutional rules, which are logical rules that can be extracted using a convolutional neural network (CNN), and w…

MFNet: Multi-feature convolutional neural network for high-density crowd counting

2020

The crowd counting task involves the issue of security, so now more and more people are concerned about it. At present, the most difficult problem of population counting consists in: how to make the model distinguish human head features more finely in the densely populated area, such as head overlap and how to find a small-scale local head feature in an image with a wide range of population density. Facing these challenges, we propose a network for multiple feature convolutional neural network, which is called MFNet. It aims to get high-quality density maps in the high-density crowd scene, and at the same time to perform the task of the count and estimation of the crowd. In terms of crowd c…

ES1D: A Deep Network for EEG-Based Subject Identification

2017

Security systems are starting to meet new technologies and new machine learning techniques, and a variety of methods to identify individuals from physiological signals have been developed. In this paper, we present ESID, a deep learning approach to identify subjects from electroencephalogram (EEG) signals captured by using a low cost device. The system consists of a Convolutional Neural Network (CNN), which is fed with the power spectral density of different EEG recordings belonging to different individuals. The network is trained for a period of one million iterations, in order to learn features related to local patterns in the spectral domain of the original signal. The performance of the…

Assessment of tumor-infiltrating TCRV γ 9V δ 2 γδ lymphocyte abundance by deconvolution of human cancers microarrays

2017

Most human blood γδ cells are cytolytic TCRVγ9Vδ2+lymphocytes with antitumor activity. They are currently investigated in several clinical trials of cancer immunotherapy but so far, their tumor infiltration has not been systematically explored across human cancers. Novel algorithms allowing the deconvolution of bulk tumor transcriptomes to find the relative proportions of infiltrating leucocytes, such as CIBERSORT, should be appropriate for this aim but in practice they fail to accurately recognize γδ T lymphocytes. Here, by implementing machine learning from microarray data, we first improved the computational identification of blood-derived TCRVγ9Vδ2+γδ lymphocytes and then appl…

Two simple criteria to estimate an objective's performance when imaging in non design tissue clearing solutions

2019

Tissue clearing techniques are undergoing a renaissance motivated by the need to image fluorescent neurons, and other cells, deep in the sample without physical sectioning. Optical transparency is achieved by equilibrating tissues with high refractive index (RI) solutions. When the microscope objective is not perfectly matched to the RI of the cleared sample, aberrations are introduced. We present two simple-to-calculate numerical criteria predicting: (i) the degradation in image quality (brightness and resolution) from optimal conditions of any clearing solution/objective combination; (ii) which objective, among several available, achieves the highest resolution in a given medium. We deriv…

Deep Learning Architectures for DNA Sequence Classification

2017

DNA sequence classification is a key task in a generic computational framework for biomedical data analysis, and in recent years several machine learning technique have been adopted to successful accomplish with this task. Anyway, the main difficulty behind the problem remains the feature selection process. Sequences do not have explicit features, and the commonly used representations introduce the main drawback of the high dimensionality. For sure, machine learning method devoted to supervised classification tasks are strongly dependent on the feature extraction step, and in order to build a good representation it is necessary to recognize and measure meaningful details of the items to cla…