A robust interclass and intraclass loss function for deep learning based tongue segmentation

2020 ◽  
Vol 32 (22) ◽  
Author(s):  
Yuanzheng Cai ◽  
Tao Wang ◽  
Wei Liu ◽  
Zhiming Luo
Keyword(s):  
Deep Learning ◽  
Loss Function ◽  
Tongue Segmentation

scholarly journals Study on Radar Echo-Filling in an Occlusion Area by a Deep Learning Algorithm

2021 ◽  
Vol 13 (9) ◽  
pp. 1779
Author(s):  
Xiaoyan Yin ◽  
Zhiqun Hu ◽  
Jiafeng Zheng ◽  
Boyong Li ◽  
Yuanyuan Zuo
Keyword(s):  
Deep Learning ◽  
Loss Function ◽  
Learning Algorithm ◽  
Weather Radar ◽  
Loss Functions ◽  
Training Dataset ◽  
Echo Intensity ◽  
Common Mean ◽  
Deep Learning Algorithm ◽  
Radar Beam

Radar beam blockage is an important error source that affects the quality of weather radar data. An echo-filling network (EFnet) is proposed based on a deep learning algorithm to correct the echo intensity under the occlusion area in the Nanjing S-band new-generation weather radar (CINRAD/SA). The training dataset is constructed by the labels, which are the echo intensity at the 0.5° elevation in the unblocked area, and by the input features, which are the intensity in the cube including multiple elevations and gates corresponding to the location of bottom labels. Two loss functions are applied to compile the network: one is the common mean square error (MSE), and the other is a self-defined loss function that increases the weight of strong echoes. Considering that the radar beam broadens with distance and height, the 0.5° elevation scan is divided into six range bands every 25 km to train different models. The models are evaluated by three indicators: explained variance (EVar), mean absolute error (MAE), and correlation coefficient (CC). Two cases are demonstrated to compare the effect of the echo-filling model by different loss functions. The results suggest that EFnet can effectively correct the echo reflectivity and improve the data quality in the occlusion area, and there are better results for strong echoes when the self-defined loss function is used.

A 2020 perspective on “DeRec: A data-driven approach to accurate recommendation with deep learning and weighted loss function”

2021 ◽  
Vol 48 ◽  
pp. 101064
Author(s):  
Wen Zhang ◽  
Qiang Wang ◽  
Ye Yang ◽  
Taketoshi Yoshida
Keyword(s):  
Deep Learning ◽  
Loss Function ◽  
Data Driven ◽  
Data Driven Approach

scholarly journals Assessing the Impact of the Loss Function, Architecture and Image Type for Deep Learning-Based Wildfire Segmentation

2021 ◽  
Vol 11 (15) ◽  
pp. 7046
Author(s):  
Jorge Francisco Ciprián-Sánchez ◽  
Gilberto Ochoa-Ruiz ◽  
Lucile Rossi ◽  
Frédéric Morandini
Keyword(s):  
Deep Learning ◽  
Loss Function ◽  
State Of The Art ◽  
Fire Detection ◽  
Loss Functions ◽  
Wildfire Spread ◽  
Combine Information ◽  
The Impact ◽  
Image Type ◽  
Segmentation Models

Wildfires stand as one of the most relevant natural disasters worldwide, particularly more so due to the effect of climate change and its impact on various societal and environmental levels. In this regard, a significant amount of research has been done in order to address this issue, deploying a wide variety of technologies and following a multi-disciplinary approach. Notably, computer vision has played a fundamental role in this regard. It can be used to extract and combine information from several imaging modalities in regard to fire detection, characterization and wildfire spread forecasting. In recent years, there has been work pertaining to Deep Learning (DL)-based fire segmentation, showing very promising results. However, it is currently unclear whether the architecture of a model, its loss function, or the image type employed (visible, infrared, or fused) has the most impact on the fire segmentation results. In the present work, we evaluate different combinations of state-of-the-art (SOTA) DL architectures, loss functions, and types of images to identify the parameters most relevant to improve the segmentation results. We benchmark them to identify the top-performing ones and compare them to traditional fire segmentation techniques. Finally, we evaluate if the addition of attention modules on the best performing architecture can further improve the segmentation results. To the best of our knowledge, this is the first work that evaluates the impact of the architecture, loss function, and image type in the performance of DL-based wildfire segmentation models.

scholarly journals Deep Learning of Appearance Affinity for Multi-Object Tracking and Re-Identification: A Comparative View

Electronics ◽  
2020 ◽  
Vol 9 (11) ◽  
pp. 1757
Author(s):  
María J. Gómez-Silva ◽  
Arturo de la Escalera ◽  
José M. Armingol
Keyword(s):  
Deep Learning ◽  
Object Tracking ◽  
Loss Function ◽  
Neural Model ◽  
Training Data ◽  
Learning Approaches ◽  
The Core ◽  
Triplet Loss ◽  
Affinity Model

Recognizing the identity of a query individual in a surveillance sequence is the core of Multi-Object Tracking (MOT) and Re-Identification (Re-Id) algorithms. Both tasks can be addressed by measuring the appearance affinity between people observations with a deep neural model. Nevertheless, the differences in their specifications and, consequently, in the characteristics and constraints of the available training data for each one of these tasks, arise from the necessity of employing different learning approaches to attain each one of them. This article offers a comparative view of the Double-Margin-Contrastive and the Triplet loss function, and analyzes the benefits and drawbacks of applying each one of them to learn an Appearance Affinity model for Tracking and Re-Identification. A batch of experiments have been conducted, and their results support the hypothesis concluded from the presented study: Triplet loss function is more effective than the Contrastive one when an Re-Id model is learnt, and, conversely, in the MOT domain, the Contrastive loss can better discriminate between pairs of images rendering the same person or not.

A novel loss function of deep learning in wind speed forecasting

Energy ◽  
2021 ◽  
pp. 121808
Author(s):  
Xi Chen ◽  
Ruyi Yu ◽  
Sajid Ullah ◽  
Dianming Wu ◽  
Zhiqiang Li ◽  
...  
Keyword(s):  
Deep Learning ◽  
Wind Speed ◽  
Loss Function ◽  
Wind Speed Forecasting

A novel enhanced softmax loss function for brain tumour detection using deep learning

2020 ◽  
Vol 330 ◽  
pp. 108520 ◽  
Author(s):  
Sunil Maharjan ◽  
Abeer Alsadoon ◽  
P.W.C. Prasad ◽  
Thair Al-Dalain ◽  
Omar Hisham Alsadoon
Keyword(s):  
Deep Learning ◽  
Brain Tumour ◽  
Loss Function ◽  
Tumour Detection

scholarly journals HPS-Net: Multi-Task Network for Medical Image Segmentation with Predictable Performance

Symmetry ◽  
2021 ◽  
Vol 13 (11) ◽  
pp. 2107
Author(s):  
Xin Wei ◽  
Huan Wan ◽  
Fanghua Ye ◽  
Weidong Min
Keyword(s):  
Image Segmentation ◽  
Deep Learning ◽  
Loss Function ◽  
Network Structure ◽  
Medical Diagnosis ◽  
Medical Image ◽  
Medical Image Segmentation ◽  
Training Mode ◽  
Cooperative Training

In recent years, medical image segmentation (MIS) has made a huge breakthrough due to the success of deep learning. However, the existing MIS algorithms still suffer from two types of uncertainties: (1) the uncertainty of the plausible segmentation hypotheses and (2) the uncertainty of segmentation performance. These two types of uncertainties affect the effectiveness of the MIS algorithm and then affect the reliability of medical diagnosis. Many studies have been done on the former but ignore the latter. Therefore, we proposed the hierarchical predictable segmentation network (HPS-Net), which consists of a new network structure, a new loss function, and a cooperative training mode. According to our knowledge, HPS-Net is the first network in the MIS area that can generate both the diverse segmentation hypotheses to avoid the uncertainty of the plausible segmentation hypotheses and the measure predictions about these hypotheses to avoid the uncertainty of segmentation performance. Extensive experiments were conducted on the LIDC-IDRI dataset and the ISIC2018 dataset. The results show that HPS-Net has the highest Dice score compared with the benchmark methods, which means it has the best segmentation performance. The results also confirmed that the proposed HPS-Net can effectively predict TNR and TPR.

An Improved 2DCNN With Focal Loss Function for Blade Icing Detection of Wind Turbines Under Imbalanced SCADA Data

2021 ◽  
Author(s):  
Dandan Peng ◽  
Chenyu Liu ◽  
Wim Desmet ◽  
Konstantinos Gryllias
Keyword(s):  
Power Generation ◽  
Deep Learning ◽  
Wind Turbine ◽  
Loss Function ◽  
Wind Turbines ◽  
Cold Climate ◽  
Detection Methods ◽  
Physical Parameters ◽  
Learning Abilities ◽  
Imbalance Problem

Abstract The deployment of wind power plants in cold climate becomes ever more attractive due to the increased air density resulting from low temperatures, the high wind speeds, and the low population density. However, the cold climate conditions bring some additional challenges as itt can easily cause wind turbine blades to freeze. The frizzing ice on blades not only increases the energy required for the rotation of blades, resulting in a reduction in the power generation, but also increases the amplitude of the blades’ vibrations, which may cause the blade to break, affecting the power generation performance of the wind turbine and poses a threat to its safe operation. Current published blade icing detection methods focus on studying the blade icing mechanism, building the model and then judging if it is iced or not. These models vary with different wind turbines and working conditions, so expertise knowledge is required. However, deep learning techniques may solve the abovementioned problem based on their excellent feature learning abilities but until now, there are only few studies on wind turbine blade icing detection based on the deep learning technology. Therefore, this paper proposes a novel blade icing detection model, named two-dimensional convolutional neural network with focal loss function (FL-2DCNN). The network takes the raw data collected by the Supervisory Control and Data Acquisition (SCADA) system as input, automatically learns the correlation between the different physical parameters in the dataset, and captures the abnormal information, in order to accurately output the detection results. However, the amount of normal data collected by SCADA systems is usually much larger than the one of blade icing fault data, leading to a serious data imbalance problem. This problem makes it difficult for the network to obtain enough features related to the blade icing fault. Therefore the focal loss function is introduced to the FL-2DCNN to solve the aforementioned data imbalanced problem. The focal loss function can effectively balance the importance of normal samples and icing fault samples, so that the network can obtain more icing-related feature information from the icing fault samples, and thus the detection ability of the network can be improved. The experimental results of the proposed FL-2DCNN based on real SCADA data of wind turbines show that the proposed FL-2DCNN can effectively solve the sample imbalance problem and has significant potential in the blade icing detection task compared with other deep learning methods.

A novel deep learning system for facial feature extraction by fusing CNN and MB-LBP and using enhanced loss function

2020 ◽  
Vol 79 (41-42) ◽  
pp. 31027-31047
Author(s):  
Raj Silwal ◽  
Abeer Alsadoon ◽  
P. W. C. Prasad ◽  
Omar Hisham Alsadoon ◽  
Ammar Al-Qaraghuli
Keyword(s):  
Feature Extraction ◽  
Deep Learning ◽  
Loss Function ◽  
Facial Feature ◽  
Learning System ◽  
Facial Feature Extraction
Sign in / Sign up

Export Citation Format

Share Document