A survey and performance evaluation of deep learning methods for small object detection

Liu, Yang; Sun, Peng; Wergeles, Nickolas M.; Shang, Yi

doi:10.1016/j.eswa.2021.114602

Cited by 322 publications

(144 citation statements)

References 69 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Dense-sparse-dense logic is used to reduce the number of parameters. There are dense blocks similar to the Resnet model (Liu et al, 2021). These blocks are shown in figure 7.…”

Section: Cnn and Deep Learning Modelsmentioning

confidence: 95%

Comparison of deep learning models in terms of multiple object detection on satellite images

Doğan

Türkoğlu²

2021

JER is an international, peer-reviewed journal that publishes f

View full text Add to dashboard Cite

The images obtained by remote sensing contain important data about ground surface. It is an important issue to detect objects on the ground surface with these images. Deep learning models are known to give better results in studies on object detection. However, the superiority of the deep learning models over each other is unknown. For this reason, it should be clarified which model is superior in terms of object detection and which model should be used in studies. In this study, it was aimed to reveal the superiorities of deep learning models by comparing their performance in detecting multiple objects. By using 11 deep learning models that are frequently encountered in the literature, the application of detecting objects of 14 classes in the DOTA dataset were made. 49,053 objects in 888 images were used for training by using AlexNet, Vgg16, Vgg19, GoogleNet, SequezeeNet, Resnet18, Resnet50, Resnet101, Inceptionresnetv2, inceptionv3, DenseNet201 models. After the training, 13,772 objects consisting of 14 classes in 277 images were used for testing with RCNN, which is one of the object detection methods. The performance of each algorithm in 14 classes has been demonstrated by using Average Precision (AP) and Mean Average Precision (mAP) to measure the performance of the models from their metrics. In a particular class of each deep learning model, difference in performance was observed The model with the highest performance varies in each class. In the application, the most successful average mAP value of 14 classes was Vgg16 with 24.64, while the lowest was InceptionResnetV2 with 11.78. In this article, the success of deep learning models in detecting multiple objects has been demonstrated practically and it is thought to be an important resource for researchers who will study on this subject.

show abstract

“…Dense-sparse-dense logic is used to reduce the number of parameters. There are dense blocks similar to the Resnet model (Liu et al, 2021). These blocks are shown in figure 7.…”

Section: Cnn and Deep Learning Modelsmentioning

confidence: 95%

Comparison of deep learning models in terms of multiple object detection on satellite images

Doğan

Türkoğlu²

2021

JER is an international, peer-reviewed journal that publishes f

View full text Add to dashboard Cite

show abstract

“…• Light weight systems: Modern state-of-the-art graphical page object detection methods are not efficient at all. However, there is a growing need to build intelligent information extraction systems that can work effortlessly on mobile devices [64,65].…”

Section: Future Workmentioning

confidence: 99%

A Survey of Graphical Page Object Detection with Deep Neural Networks

et al. 2021

View full text Add to dashboard Cite

In any document, graphical elements like tables, figures, and formulas contain essential information. The processing and interpretation of such information require specialized algorithms. Off-the-shelf OCR components cannot process this information reliably. Therefore, an essential step in document analysis pipelines is to detect these graphical components. It leads to a high-level conceptual understanding of the documents that make the digitization of documents viable. Since the advent of deep learning, deep learning-based object detection performance has improved many folds. This work outlines and summarizes the deep learning approaches for detecting graphical page objects in document images. Therefore, we discuss the most relevant deep learning-based approaches and state-of-the-art graphical page object detection in document images. This work provides a comprehensive understanding of the current state-of-the-art and related challenges. Furthermore, we discuss leading datasets along with the quantitative evaluation. Moreover, it discusses briefly the promising directions that can be utilized for further improvements.

show abstract

“…With the development of artificial intelligence technology, deep learning is a branch of machine learning that can automatically extract features to represent the characteristics of the data [ 28 , 29 , 30 , 31 , 32 , 33 ]. In this study, we estimate the tool wear and surface roughness using a one-dimensional (1D) convolutional neural network (1D-CNN) with sensor fusion.…”

Section: Introductionmentioning

confidence: 99%

Estimation of Tool Wear and Surface Roughness Development Using Deep Learning and Sensors Fusion

Huang

Lee

2021

Sensors

View full text Add to dashboard Cite

This paper proposes an estimation approach for tool wear and surface roughness using deep learning and sensor fusion. The one-dimensional convolutional neural network (1D-CNN) is utilized as the estimation model with X- and Y-coordinate vibration signals and sound signal fusion using sensor influence analysis. First, machining experiments with computer numerical control (CNC) parameters are designed using a uniform experimental design (UED) method to guarantee the variety of collected data. The vibration, sound, and spindle current signals are collected and labeled according to the machining parameters. To speed up the degree of tool wear, an accelerated experiment is designed, and the corresponding tool wear and surface roughness are measured. An influential sensor selection analysis is proposed to preserve the estimation accuracy and to minimize the number of sensors. After sensor selection analysis, the sensor signals with better estimation capability are selected and combined using the sensor fusion method. The proposed estimation system combined with sensor selection analysis performs well in terms of accuracy and computational effort. Finally, the proposed approach is applied for on-line monitoring of tool wear with an alarm, which demonstrates the effectiveness of our approach.

show abstract

A survey and performance evaluation of deep learning methods for small object detection

Cited by 322 publications

References 69 publications

Comparison of deep learning models in terms of multiple object detection on satellite images

Comparison of deep learning models in terms of multiple object detection on satellite images

A Survey of Graphical Page Object Detection with Deep Neural Networks

Estimation of Tool Wear and Surface Roughness Development Using Deep Learning and Sensors Fusion

Contact Info

Product

Resources

About