scholarly journals MOHE-NET: MONOCULAR OBJECT HEIGHT ESTIMATION NETWORK USING DEEP LEARNING AND SCENE GEOMETRY

2021 ◽  
Vol XLIII-B2-2021 ◽  
pp. 557-564
Author(s):  
J. Wei ◽  
J. Jiang ◽  
A. Yilmaz
Keyword(s):  
Lower Cost ◽  
State Of The Art ◽  
Activation Function ◽  
Detection Task ◽  
Field Of View ◽  
Laser Ranging ◽  
Multi Layer Perceptron ◽  
Height Estimation ◽  
Set Up ◽  
Mlp Model

Abstract. Estimating the heights of objects in the field of view has applications in many tasks such as robotics, autonomous platforms and video surveillance. Object height is a concrete and indispensable characteristic people or machine could learn and capture. Many actions such as vehicle avoiding obstacles will be taken based on it. Traditionally, object height can be estimated using laser ranging, radar or stereo camera. Depending on the application, cost of these techniques may inhibit their use, especially in autonomous platforms. Use of available sensors with lower cost would make the adoption of such techniques at higher rates. Our approach to height estimation requires only a single 2D image. To solve this problem we introduce the Monocular Object Height Estimation Network (MOHE-Net) that includes a cascade of two networks. The first network performs the object detection task. This network detects the bounding box of objects of interest. This information is then input to a second network to estimate the object height and is a linear Multi-layer Perceptron (MLP). The linear MLP model models the camera-scene geometry and does not require training or contain activation function as normal MLP did. The developed approach works for static camera set up as well as moving platform. The proposed approach performs state-of-the-art and can be deployed for obstacle avoidance on autonomous platforms. Our code is available at https://github.com/OSUPCVLab/Ford2019/tree/master/Moving%20Object%20Height% 20Estimation%20Network

Operational Aspects of National Securities Depository Limited (NSDL) in Depository System

2018 ◽  
Vol 6 (8) ◽  
pp. 249
Author(s):  
Rajnikant Kumar
Keyword(s):  
Capital Markets ◽  
Service Providers ◽  
State Of The Art ◽  
Stock Exchanges ◽  
Saving Account ◽  
Business Partners ◽  
Paper Work ◽  
Set Up ◽  
Buy Back ◽  
Operational Aspects

NSDL was registered by the SEBI on June 7, 1996 as India’s first depository to facilitate trading and settlement of securities in the dematerialized form. NSDL has been set up to cater to the demanding needs of the Indian capital markets. NSDL commenced operations on November 08, 1996. NSDL has been promoted by a number of companies, the prominent of them being IDBI, UTI, NSE, SBI, HDFC Bank Ltd., etc. The initial paid up capital of NSDL was Rs. 105 crore which was reduced to Rs. 80 crore. During 2000-2001 through buy-back programme by buying back 2.5 crore shares @ 12 Rs./share. It was done to bring the size of its capital in better alignment with its financial operations and to provide same return to shareholders by gainfully deploying the excess cash available with NSDL. NSDL carries out its activities through service providers such as depository participants (DPs), issuing companies and their registrars and share transfer agents and clearing corporations/ clearing houses of stock exchanges. These entities are NSDL's business partners and are integrated in to the NSDL depository system to provide various services to investors and clearing members. The investor can get depository services through NSDL's depository participants. An investor needs to open a depository account with a depository participant to avail of depository facilities. Depository system essentially aims at eliminating the voluminous and cumbersome paper work involved in the scrip-based system and offers scope for ‘paperless’ trading through state-of-the-art technology. A depository can be compared to a bank. A depository holds securities of investors in the form of electronic accounts, in the same way as bank holds money in a saving account. Besides, holding securities, a depository also provides services related to transactions in securities.

scholarly journals Extrinsic Camera Calibration with Line-Laser Projection

Sensors ◽  
2021 ◽  
Vol 21 (4) ◽  
pp. 1091
Author(s):  
Izaak Van Crombrugge ◽  
Rudi Penne ◽  
Steve Vanlanduit
Keyword(s):  
Camera Calibration ◽  
Real World ◽  
Large Scale ◽  
State Of The Art ◽  
Bundle Adjustment ◽  
Field Of View ◽  
Extrinsic Calibration ◽  
Practical Procedure ◽  
Partial Overlap

Knowledge of precise camera poses is vital for multi-camera setups. Camera intrinsics can be obtained for each camera separately in lab conditions. For fixed multi-camera setups, the extrinsic calibration can only be done in situ. Usually, some markers are used, like checkerboards, requiring some level of overlap between cameras. In this work, we propose a method for cases with little or no overlap. Laser lines are projected on a plane (e.g., floor or wall) using a laser line projector. The pose of the plane and cameras is then optimized using bundle adjustment to match the lines seen by the cameras. To find the extrinsic calibration, only a partial overlap between the laser lines and the field of view of the cameras is needed. Real-world experiments were conducted both with and without overlapping fields of view, resulting in rotation errors below 0.5°. We show that the accuracy is comparable to other state-of-the-art methods while offering a more practical procedure. The method can also be used in large-scale applications and can be fully automated.

scholarly journals BeautyNet: Joint Multiscale CNN and Transfer Learning Method for Unconstrained Facial Beauty Prediction

2019 ◽  
Vol 2019 ◽  
pp. 1-14 ◽  
Author(s):  
Yikui Zhai ◽  
He Cao ◽  
Wenbo Deng ◽  
Junying Gan ◽  
Vincenzo Piuri ◽  
...  
Keyword(s):  
Transfer Learning ◽  
Classification Accuracy ◽  
Learning Strategy ◽  
State Of The Art ◽  
Activation Function ◽  
Training Data ◽  
Fine Grained ◽  
Pattern Recognition Problem ◽  
Face Features ◽  
Facial Beauty

Because of the lack of discriminative face representations and scarcity of labeled training data, facial beauty prediction (FBP), which aims at assessing facial attractiveness automatically, has become a challenging pattern recognition problem. Inspired by recent promising work on fine-grained image classification using the multiscale architecture to extend the diversity of deep features, BeautyNet for unconstrained facial beauty prediction is proposed in this paper. Firstly, a multiscale network is adopted to improve the discriminative of face features. Secondly, to alleviate the computational burden of the multiscale architecture, MFM (max-feature-map) is utilized as an activation function which can not only lighten the network and speed network convergence but also benefit the performance. Finally, transfer learning strategy is introduced here to mitigate the overfitting phenomenon which is caused by the scarcity of labeled facial beauty samples and improves the proposed BeautyNet’s performance. Extensive experiments performed on LSFBD demonstrate that the proposed scheme outperforms the state-of-the-art methods, which can achieve 67.48% classification accuracy.

scholarly journals Particle Size Estimation in Mixed Commercial Waste Images Using Deep Learning

2021 ◽  
Author(s):  
Phongsathorn Kittiworapanya ◽  
Kitsuchart Pasupa ◽  
Peter Auer
Keyword(s):  
Computer Vision ◽  
Particle Size ◽  
Deep Learning ◽  
Waste Management ◽  
State Of The Art ◽  
Learning Algorithms ◽  
Input Image ◽  
Size Estimation ◽  
Waste Particles ◽  
Set Up

<div>We assessed several state-of-the-art deep learning algorithms and computer vision techniques for estimating the particle size of mixed commercial waste from images. In waste management, the first step is often coarse shredding, using the particle size to set up the shredder machine. The difficulty is separating the waste particles in an image, which can not be performed well. This work focused on estimating size by using the texture from the input image, captured at a fixed height from the camera lens to the ground. We found that EfficientNet achieved the best performance of 0.72 on F1-Score and 75.89% on accuracy.<br></div>

scholarly journals Multi Layer Perceptron Modelling in the Housing Market

2020 ◽  
Author(s):  
Ku Ruhana Ku Mahamud ◽  
Azuraliza Abu Bakar ◽  
Norita Md. Norwawi
Keyword(s):  
Regression Analysis ◽  
Real Estate ◽  
Housing Market ◽  
Predictive Ability ◽  
Multi Layer Perceptron ◽  
Kuala Lumpur ◽  
Mlp Model

The study examines the use of multi layer perception network (MLP) in predicting the price of terrace houses in Kuala Lumpur (KL). Nine factors that significantly influence the price were used in this attempt. Housing data from 1994 to 1996 were presented to the network for training. Tested results from the model obtained for various years were compared using regression analysis. The study provides the predictive ability of the trained MLP model that can be used as an alternative predictor in real estate analysis.  

Sparse Autoencoder-Based Multi-Head Deep Neural Networks for Machinery Fault Diagnostics With Novelty Detection

2020 ◽  
Author(s):  
Zhe Yang ◽  
Dejan Gjorgjevikj ◽  
Jian-Yu Long ◽  
Yan-Yang Zi ◽  
Shao-Hui Zhang ◽  
...  
Keyword(s):  
Supervised Classification ◽  
Deep Neural Networks ◽  
State Of The Art ◽  
Novelty Detection ◽  
Activation Function ◽  
Reconstruction Error ◽  
Fine Tuning ◽  
Sparse Autoencoder

Abstract Novelty detection is a challenging task for the machinery fault diagnosis. A novel fault diagnostic method is developed for dealing with not only diagnosing the known type of defect, but also detecting novelties, i.e. the occurrence of new types of defects which have never been recorded. To this end, a sparse autoencoder-based multi-head Deep Neural Network (DNN) is presented to jointly learn a shared encoding representation for both unsupervised reconstruction and supervised classification of the monitoring data. The detection of novelties is based on the reconstruction error. Moreover, the computational burden is reduced by directly training the multi-head DNN with rectified linear unit activation function, instead of performing the pre-training and fine-tuning phases required for classical DNNs. The addressed method is applied to a benchmark bearing case study and to experimental data acquired from a delta 3D printer. The results show that it is able to accurately diagnose known types of defects, as well as to detect unknown defects, outperforming other state-of-the-art methods.

scholarly journals Low-cost high-efficiency system for solar-driven conversion of CO2 to hydrocarbons

2019 ◽  
Vol 116 (20) ◽  
pp. 9735-9740 ◽  
Author(s):  
Tran Ngoc Huan ◽  
Daniel Alves Dalla Corte ◽  
Sarah Lamaison ◽  
Dilan Karapinar ◽  
Lukas Lutz ◽  
...  
Keyword(s):  
High Efficiency ◽  
Electrochemical Cell ◽  
State Of The Art ◽  
Low Cost ◽  
Photovoltaic Cells ◽  
Energy Losses ◽  
Earth Abundant ◽  
Set Up ◽  
System Coupling ◽  
Perovskite Photovoltaic

Conversion of carbon dioxide into hydrocarbons using solar energy is an attractive strategy for storing such a renewable source of energy into the form of chemical energy (a fuel). This can be achieved in a system coupling a photovoltaic (PV) cell to an electrochemical cell (EC) for CO2 reduction. To be beneficial and applicable, such a system should use low-cost and easily processable photovoltaic cells and display minimal energy losses associated with the catalysts at the anode and cathode and with the electrolyzer device. In this work, we have considered all of these parameters altogether to set up a reference PV–EC system for CO2 reduction to hydrocarbons. By using the same original and efficient Cu-based catalysts at both electrodes of the electrolyzer, and by minimizing all possible energy losses associated with the electrolyzer device, we have achieved CO2 reduction to ethylene and ethane with a 21% energy efficiency. Coupled with a state-of-the-art, low-cost perovskite photovoltaic minimodule, this system reaches a 2.3% solar-to-hydrocarbon efficiency, setting a benchmark for an inexpensive all–earth-abundant PV–EC system.

scholarly journals Comparing Activation Functions in Modeling Shoreline Variation Using Multilayer Perceptron Neural Network

Water ◽  
2020 ◽  
Vol 12 (5) ◽  
pp. 1281
Author(s):  
Je-Chian Chen ◽  
Yu-Min Wang
Keyword(s):  
Neural Network ◽  
Multilayer Perceptron ◽  
Activation Function ◽  
Aerial Survey ◽  
Activation Functions ◽  
Shoreline Changes ◽  
Extensive Evaluation ◽  
Aerial Vehicle ◽  
Sine Functions ◽  
Mlp Model

The study has modeled shoreline changes by using a multilayer perceptron (MLP) neural network with the data collected from five beaches in southern Taiwan. The data included aerial survey maps of the Forestry Bureau for years 1982, 2002, and 2006, which served as predictors, while the unmanned aerial vehicle (UAV) surveyed data of 2019 served as the respondent. The MLP was configured using five different activation functions with the aim of evaluating their significance. These functions were Identity, Tahn, Logistic, Exponential, and Sine Functions. The results have shown that the performance of an MLP model may be affected by the choice of an activation function. Logistic and the Tahn activation functions outperformed the other models, with Logistic performing best in three beaches and Tahn having the rest. These findings suggest that the application of machine learning to shoreline changes should be accompanied by an extensive evaluation of the different activation functions.

scholarly journals Fast and Deep Graph Neural Networks

2020 ◽  
Vol 34 (04) ◽  
pp. 3898-3905 ◽  
Author(s):  
Claudio Gallicchio ◽  
Alessio Micheli
Keyword(s):  
Neural Network ◽  
Dynamical System ◽  
Neural Networks ◽  
State Of The Art ◽  
Input Graph ◽  
Sparse Networks ◽  
Set Up ◽  
The Stability ◽  
Graph Neural Networks ◽  
Architectural Organization

We address the efficiency issue for the construction of a deep graph neural network (GNN). The approach exploits the idea of representing each input graph as a fixed point of a dynamical system (implemented through a recurrent neural network), and leverages a deep architectural organization of the recurrent units. Efficiency is gained by many aspects, including the use of small and very sparse networks, where the weights of the recurrent units are left untrained under the stability condition introduced in this work. This can be viewed as a way to study the intrinsic power of the architecture of a deep GNN, and also to provide insights for the set-up of more complex fully-trained models. Through experimental results, we show that even without training of the recurrent connections, the architecture of small deep GNN is surprisingly able to achieve or improve the state-of-the-art performance on a significant set of tasks in the field of graphs classification.

Sign in / Sign up

Export Citation Format

Share Document