scholarly journals Intelligent recommendation algorithm of mobile application crowdsourcing test based on deep learning

2021 ◽  
Vol 39 (5) ◽  
pp. 1049-1056
Author(s):  
Jing CHENG ◽  
Wei WANG ◽  
Zhengyi SHUAI
Keyword(s):  
Deep Learning ◽  
Mobile Applications ◽  
Learning Ability ◽  
Test Quality ◽  
Recommendation Algorithm ◽  
Test Task ◽  
Training Time ◽  
Deep Feature ◽  
Task Requirements ◽  
And Training

As the functions of mobile applications become more and more complex, the crowdsourcing testing puts higher demands on the professional skills of testers. Therefore, it is an important factor to ensure test quality how to effectively match test task requirements with test personnel's skill level and achieve accurate crowdsourcing test task recommendation. This paper proposes a crowdsourcing test task recommendation algorithm for mobile applications based on deep learning. Firstly, feature analysis is carried out for testing tasks and testers, and feature systems are designed respectively. Second, the resulting characteristic data is used as input data for the Stacked Marginalized Denoising Autoencoder (SMDA). The deep feature data learned from SMDA are combined as the input of Deep Neural Networks (DNN). Finally, the learning ability of DNN is used for prediction. Experimental results show that the proposed algorithm has obvious advantages in both performance and training time compared with CDL and AUTOSVD ++, which verifies the effectiveness of the proposed algorithm. The proposed algorithm can recommend testing tasks to appropriate testers and improve the precision of the algorithm.

scholarly journals Mosquito Larvae Detection using Deep Learning

2019 ◽  
Vol 8 (12) ◽  
pp. 804-809
Keyword(s):  
Deep Learning ◽  
Mobile Application ◽  
Web Application ◽  
Individual Performance ◽  
Mosquito Larvae ◽  
Accuracy Score ◽  
Performance Category ◽  
Training Time ◽  
The Cost ◽  
And Training

Dengue cases has become endemic in Malaysia. The cost of operation to exterminate mosquito habitats are also high. To do effective operation, information from community are crucial. But, without knowing the characteristic of Aedes larvae it is hard to recognize the larvae without guide from the expert. The use of deep learning in image classification and recognition is crucial to tackle this problem. The purpose of this project is to conduct a study of characteristics of Aedes larvae and determine the best convolutional neural network model in classifying the mosquito larvae. 3 performance evaluation vector which is accuracy, log-loss and AUC-ROC will be used to measure the model’s individual performance. Then performance category which consist of Accuracy Score, Loss Score, File Size Score and Training Time Score will be used to evaluate which model is the best to be implemented into web application or mobile application. From the score collected for each model, ResNet50 has proved to be the best model in classifying the mosquito larvae species.

scholarly journals A Novel Deep Feature Learning Method Based on the Fused-Stacked AEs for Planetary Gear Fault Diagnosis

Energies ◽  
2019 ◽  
Vol 12 (23) ◽  
pp. 4522
Author(s):  
Xihui Chen ◽  
Aimin Ji ◽  
Gang Cheng
Keyword(s):  
Deep Learning ◽  
Fault Diagnosis ◽  
Feature Learning ◽  
Planetary Gear ◽  
Learning Ability ◽  
Energy Industry ◽  
Deep Feature ◽  
Gear Fault ◽  
Deep Feature Learning ◽  
Electromechanical Equipment

Planetary gear is the key component of the transmission system of electromechanical equipment for energy industry, and it is easy to damage, which affects the reliability and operation efficiency of electromechanical equipment of energy industry. Therefore, it is of great significance to extract the useful fault features and diagnose faults based on raw vibration signals. In this paper, a novel deep feature learning method based on the fused-stacked autoencoders (AEs) for planetary gear fault diagnosis was proposed. First, to improve the data learning ability and the robustness of feature extraction process of AE model, the sparse autoencoder (SAE) and the contractive autoencoder (CAE) were studied, respectively. Then, the quantum ant colony algorithm (QACA) was used to optimize the specific location and key parameters of SAEs and CAEs in deep learning architecture, and multiple SAEs and multiple CAEs were stacked alternately to form a novel deep learning architecture, which gave the deep learning architecture better data learning ability and robustness of feature extraction. The experimental results show that the proposed method can address the raw vibration signals of planetary gear. Compared with other deep learning architectures and shallow learning architecture, the proposed method has better diagnosis performance, and it is an effective method of deep feature learning and fault diagnosis.

scholarly journals Influencing factors analysis in pear disease recognition using deep learning

2020 ◽  
Author(s):  
Fang Yang ◽  
Fuzhong Li ◽  
Kai Zhang ◽  
Wuping Zhang ◽  
Shancang Li
Keyword(s):  
Deep Learning ◽  
Influencing Factors ◽  
Recognition Accuracy ◽  
Sample Collection ◽  
Training Time ◽  
Factors Analysis ◽  
Resolution Scheme ◽  
And Training ◽  
Different Levels ◽  
Pear Disease

AbstractInfluencing factors analysis plays an important role in plant disease identification. This paper explores the key influencing factors and severity recognition of pear diseases using deep learning based on our established pear disease database (PDD2018), which contains 4944 pieces of diseased leaves. Using the deep learning neural networks, including VGG16, Inception V3, ResNet50 and ResNet101, we developed a “DL network + resolution” scheme that can be used in influencing factors analysis and diseases recognition at six different levels. The experimental results demonstrated that the resolution is directly proportional to disease recognition accuracy and training time and the recognition accuracies for pear diseases are up to 99.44%,98.43%, and 97.67% for Septoria piricola (SP), Alternaria alternate (AA), and Gymnosporangium haracannum (GYM), respectively. The results also shown that a forward suggestion on disease sample collection can significantly reduce the false recognition accuracy.

scholarly journals Foot Depth Map Point Cloud Completion using Deep Learning with Residual Blocks

2017 ◽  
Vol 3 (1) ◽  
Author(s):  
Nolan Lunscher ◽  
John Zelek
Keyword(s):  
Deep Learning ◽  
Point Cloud ◽  
Depth Map ◽  
3D Scanning ◽  
Training Time ◽  
Shoe Size ◽  
Learning Technique ◽  
Footwear Fit ◽  
And Training

Fit is extremely important in footwear as fit largely determines performanceand comfort. Current footwear fit estimation mainly usesonly shoe size, which is extremely limited in characterizing theshape of a foot or the shape of a shoe. 3D scanning presents asolution to this, where a foot shape can be captured and virtuallyfit with shoe models. Traditional 3D scanning techniques have theirown complications however, stemming from their need to collectviews covering all aspects of an object. In this work we explore adeep learning technique to compete a foot scan point cloud frominformation contained in a single depth map view. We examine thebenefits of implementing residual blocks in architectures for this application,and find that they can improve accuracies while reducingmodel size and training time.

scholarly journals Deep Propagation Based Image Matting

2018 ◽  
Author(s):  
Yu Wang ◽  
Yi Niu ◽  
Peiyong Duan ◽  
Jianwei Lin ◽  
Yuanjie Zheng
Keyword(s):  
Deep Learning ◽  
Image Matting ◽  
Semantic Level ◽  
Learning Module ◽  
Deep Feature ◽  
Deep Image ◽  
Alpha Matte ◽  
Image Representations ◽  
Deep Feature Extraction ◽  
And Training

In this paper, we propose a deep propagation based image matting framework by introducing deep learning into learning an alpha matte propagation principal. Our deep learning architecture is a concatenation of a deep feature extraction module, an affinity learning module and a matte propagation module. These three modules are all differentiable and can be optimized jointly via an end-to-end training process. Our framework results in a semantic-level pairwise similarity of pixels for propagation by learning deep image representations adapted to matte propagation. It combines the power of deep learning and matte propagation and can therefore surpass prior state-of-the-art matting techniques in terms of both accuracy and training complexity, as validated by our experimental results from 243K images created based on two benchmark matting databases.

Plant Leaf Classification using Convolutional Neural Network

2020 ◽  
Vol 13 ◽  
Author(s):  
Nidhi ◽  
Jay Kant Pratap Singh Yadav
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Image Data ◽  
Learning Ability ◽  
Training Time ◽  
Automatic Feature Extraction ◽  
Computer Vision Applications ◽  
Shape Characteristics ◽  
And Training

Introduction: Convolutional Neural Network (CNNet) has proven the indispensable system in order to perform the recognition and classification tasks in different computer vision applications. The purpose of this study is to exploit the marvelous learning ability of CNNet in the image classification field. Method: In order to circumvent the overfitting issues and to enhance the generalization potential of the proposed FLCNNet, augmentation has been performed on the Flavia dataset that impose translation and rotation techniques to perform the augmentation with the transformed leaves having the same labels as the original ones. Both the classification models executed using; one without augmentation and one with the augmentation data are compared to check the effectiveness of the augmentation hence the aim of the proposed work. Moreover, Edge detection technique has been applied to extract the shape of the leaf images, in order to classify them accordingly. Thereafter, the FLCNNet is trained and tested for the dataset, with and without augmentation. Results: The results are gathered in terms of accuracy and training time for both datasets. The Augmented dataset (dataset 2) has been found effective and more feasible for classification without misguiding the network to learn (avoid overfitting) as compared to the dataset without augmentation (dataset 1). Conclusion: This paper proposed the Five Layer Convolution Neural Network (FLCNNet) method to classify plant leaves based on their shape. This approach can classify 8 types of leaves using automatic feature extraction, by utilizing their shape characteristics. To avoid the overfitting condition and make the performance better. We aimed to perform the classification of the augmented leaf dataset. Discussion: We proposed a five Layer CNNet (FLCNNet) to classify the leaf image data into different classes or labels based on the shape characteristics of the leaves.

scholarly journals P38 COVID-19 impact on Surgical Training And Recovery Planning (COVID-STAR) Survey

BJS Open ◽  
2021 ◽  
Vol 5 (Supplement_1) ◽  
Author(s):  
◽  
Joshua Clements
Keyword(s):  
Surgical Training ◽  
Negative Impact ◽  
Healthcare Delivery ◽  
Cross Sectional Study ◽  
Surgical Trainee ◽  
Surgical Specialty ◽  
Cross Sectional ◽  
Training Time ◽  
Surgical Trainees ◽  
And Training

Abstract Background The COVID-19 pandemic has resulted in dynamic changes to healthcare delivery. Surgery as a specialty has been significantly affected and with that the delivery of surgical training. Method This national, collaborative, cross sectional study comprising 13 surgical trainee associations distributed a pan surgical specialty survey on the COVID-19 impact on surgical training over a 4-week period (11th May - 8th June 2020). The survey was voluntary and open to medical students and surgical trainees of all specialties and training grades. All aspects of training were qualitatively assessed. This study was reported according to STROBE guidelines. Results 810 completed responses were analysed. (M401: F 390) with representation from all deaneries and training grades. 41% of respondents (n = 301) were redeployed with 74% (n = 223) redeployed > 4 weeks. Complete loss of training was reported in elective operating (69.5% n = 474), outpatient activity (67.3%, n = 457), Elective endoscopy (69.5% n = 246) with > 50% reduction in training time reported in emergency operating (48%, n = 326) and completion of work-based assessments (WBA) (46%, n = 309). 81% (n = 551) reported course cancellations and departmental and regional teaching programmes were cancelled without rescheduling in 58% and 60% of cases respectively. A perceived lack of Elective operative exposure and completions of WBA’s were the primary reported factor affecting potential training progression. Overall, > 50% of trainees (n = 377) felt they would not meet the competencies required for that training period. Conclusion This study has demonstrated a perceived negative impact on numerous aspects of surgical training affecting all training specialties and grades.

Smart paddy field monitoring system using deep learning and IoT

2021 ◽  
pp. 1063293X2198894
Author(s):  
Prabira Kumar Sethy ◽  
Santi Kumari Behera ◽  
Nithiyakanthan Kannan ◽  
Sridevi Narayanan ◽  
Chanki Pandey
Keyword(s):  
Feature Extraction ◽  
Deep Learning ◽  
Transfer Learning ◽  
Paddy Field ◽  
New Technology ◽  
Support Vector ◽  
Nitrogen Status ◽  
Deep Feature ◽  
Per Capita ◽  
Deep Feature Extraction

Paddy is an essential nutrient worldwide. Rice gives 21% of worldwide human per capita energy and 15% of per capita protein. Asia represented 60% of the worldwide populace, about 92% of the world’s rice creation, and 90% of worldwide rice utilization. With the increase in population, the demand for rice is increased. So, the productivity of farming is needed to be enhanced by introducing new technology. Deep learning and IoT are hot topics for research in various fields. This paper suggested a setup comprising deep learning and IoT for monitoring of paddy field remotely. The vgg16 pre-trained network is considered for the identification of paddy leaf diseases and nitrogen status estimation. Here, two strategies are carried out to identify images: transfer learning and deep feature extraction. The deep feature extraction approach is combined with a support vector machine (SVM) to classify images. The transfer learning approach of vgg16 for identifying four types of leaf diseases and prediction of nitrogen status results in 79.86% and 84.88% accuracy. Again, the deep features of Vgg16 and SVM results for identifying four types of leaf diseases and prediction of nitrogen status have achieved an accuracy of 97.31% and 99.02%, respectively. Besides, a framework is suggested for monitoring of paddy field remotely based on IoT and deep learning. The suggested prototype’s superiority is that it controls temperature and humidity like the state-of-the-art and can monitor the additional two aspects, such as detecting nitrogen status and diseases.

scholarly journals Copy-Move Forgery Detection (CMFD) Using Deep Learning for Image and Video Forensics

2021 ◽  
Vol 7 (3) ◽  
pp. 59
Author(s):  
Yohanna Rodriguez-Ortega ◽  
Dora M. Ballesteros ◽  
Diego Renza
Keyword(s):  
Deep Learning ◽  
Transfer Learning ◽  
Training Data ◽  
Video Editing ◽  
Forgery Detection ◽  
Copy Move Forgery Detection ◽  
The Impact ◽  
And Training ◽  
Selection Of ◽  
Traditional Image

With the exponential growth of high-quality fake images in social networks and media, it is necessary to develop recognition algorithms for this type of content. One of the most common types of image and video editing consists of duplicating areas of the image, known as the copy-move technique. Traditional image processing approaches manually look for patterns related to the duplicated content, limiting their use in mass data classification. In contrast, approaches based on deep learning have shown better performance and promising results, but they present generalization problems with a high dependence on training data and the need for appropriate selection of hyperparameters. To overcome this, we propose two approaches that use deep learning, a model by a custom architecture and a model by transfer learning. In each case, the impact of the depth of the network is analyzed in terms of precision (P), recall (R) and F1 score. Additionally, the problem of generalization is addressed with images from eight different open access datasets. Finally, the models are compared in terms of evaluation metrics, and training and inference times. The model by transfer learning of VGG-16 achieves metrics about 10% higher than the model by a custom architecture, however, it requires approximately twice as much inference time as the latter.

Sign in / Sign up

Export Citation Format

Share Document