Attentional feature pyramid network for small object detection

Min, Kyungseo; Lee, Gunhee; Lee, Seong-Whan

doi:10.1016/j.neunet.2022.08.029

Cited by 39 publications

(14 citation statements)

References 61 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The characteristic pyramid structure makes the feature map used for each layer of prediction incorporate features of different resolutions and different semantic strengths ( 16 , 17 ). Thus, it is better to detect gallbladder targets of different sizes than ResNet-101 as a feature extraction and classification model ( 18 , 19 ). Therefore, model 2 was slightly better than model 1 in terms of object extraction and classification.…”

Section: Discussionmentioning

confidence: 99%

Is the diagnostic model based on convolutional neural network superior to pediatric radiologists in the ultrasonic diagnosis of biliary atresia?

Duan,

Yang,

Zhu

et al. 2024

Front. Med.

View full text Add to dashboard Cite

BackgroundMany screening and diagnostic methods are currently available for biliary atresia (BA), but the early and accurate diagnosis of BA remains a challenge with existing methods. This study aimed to use deep learning algorithms to intelligently analyze the ultrasound image data, build a BA ultrasound intelligent diagnostic model based on the convolutional neural network, and realize an intelligent diagnosis of BA.MethodsA total of 4,887 gallbladder ultrasound images of infants with BA, non-BA hyperbilirubinemia, and healthy infants were collected. Two mask region convolutional neural network (Mask R-CNN) models based on different backbone feature extraction networks were constructed. The diagnostic performance between the two models was compared through good-quality images at the image level and the patient level. The diagnostic performance between the two models was compared through poor-quality images. The diagnostic performance of BA between the model and four pediatric radiologists was compared at the image level and the patient level.ResultsThe classification performance of BA in model 2 was slightly higher than that in model 1 in the test set, both at the image level and at the patient level, with a significant difference of p = 0.0365 and p = 0.0459, respectively. The classification accuracy of model 2 was slightly higher than that of model 1 in poor-quality images (88.3% vs. 86.4%), and the difference was not statistically significant (p = 0.560). The diagnostic performance of model 2 was similar to that of the two radiology experts at the image level, and the differences were not statistically significant. The diagnostic performance of model 2 in the test set was higher than that of the two radiology experts at the patient level (all p < 0.05).ConclusionThe performance of model 2 based on Mask R-CNN in the diagnosis of BA reached or even exceeded the level of pediatric radiology experts.

show abstract

Section: Discussionmentioning

confidence: 99%

Is the diagnostic model based on convolutional neural network superior to pediatric radiologists in the ultrasonic diagnosis of biliary atresia?

Duan,

Yang,

Zhu

et al. 2024

Front. Med.

View full text Add to dashboard Cite

show abstract

“…Many studies have integrated both approaches to augment the feature extraction capabilities for small objects. For instance, in studies [23][24][25], researchers have enhanced the network's perception of fine details on small objects by combining various attention mechanisms within the feature pyramid network, yielding promising results. However, this does not imply that a more complex model is necessarily better.…”

Section: Small Target Detectionmentioning

confidence: 99%

Mining the Micro-Trajectory of Two-Wheeled Non-Motorized Vehicles Based on the Improved YOLOx

Zhou,

Zhao,

Yang

et al. 2024

Sensors

View full text Add to dashboard Cite

Two-wheeled non-motorized vehicles (TNVs) have become the primary mode of transportation for short-distance travel among residents in many underdeveloped cities in China due to their convenience and low cost. However, this trend also brings corresponding risks of traffic accidents. Therefore, it is necessary to analyze the driving behavior characteristics of TNVs through their trajectory data in order to provide guidance for traffic safety. Nevertheless, the compact size, agile steering, and high maneuverability of these TNVs pose substantial challenges in acquiring high-precision trajectories. These characteristics complicate the tracking and analysis processes essential for understanding their movement patterns. To tackle this challenge, we propose an enhanced You Only Look Once Version X (YOLOx) model, which incorporates a median pooling-Convolutional Block Attention Mechanism (M-CBAM). This model is specifically designed for the detection of TNVs, and aims to improve accuracy and efficiency in trajectory tracking. Furthermore, based on this enhanced YOLOx model, we have developed a micro-trajectory data mining framework specifically for TNVs. Initially, the paper establishes an aerial dataset dedicated to the detection of TNVs, which then serves as a foundational resource for training the detection model. Subsequently, an augmentation of the Convolutional Block Attention Mechanism (CBAM) is introduced, integrating median pooling to amplify the model’s feature extraction capabilities. Subsequently, additional detection heads are integrated into the YOLOx model to elevate the detection rate of small-scale targets, particularly focusing on TNVs. Concurrently, the Deep Sort algorithm is utilized for the precise tracking of vehicle targets. The process culminates with the reconstruction of trajectories, which is achieved through a combination of video stabilization, coordinate mapping, and filtering denoising techniques. The experimental results derived from our self-constructed dataset reveal that the enhanced YOLOx model demonstrates superior detection performance in comparison to other analogous methods. The comprehensive framework accomplishes an average trajectory recall rate of 85% across three test videos. This significant achievement provides a reliable method for data acquisition, which is essential for investigating the micro-level operational mechanisms of TNVs. The results of this study can further contribute to the understanding and improvement of traffic safety on mixed-use roads.

show abstract

“…An FPN is a multiscale pyramid network that detects objects of varying sizes in images by using feature maps of varying resolutions [ 84 ]. An FPN has been used to recognize and segment small objects in aerial photos and to segment buildings in satellite images [ 85 , 86 , 87 ].…”

Section: Literature Reviewmentioning

confidence: 99%

YOLOv5-FPN: A Robust Framework for Multi-Sized Cell Counting in Fluorescence Images

et al. 2023

View full text Add to dashboard Cite

Cell counting in fluorescence microscopy is an essential task in biomedical research for analyzing cellular dynamics and studying disease progression. Traditional methods for cell counting involve manual counting or threshold-based segmentation, which are time-consuming and prone to human error. Recently, deep learning-based object detection methods have shown promising results in automating cell counting tasks. However, the existing methods mainly focus on segmentation-based techniques that require a large amount of labeled data and extensive computational resources. In this paper, we propose a novel approach to detect and count multiple-size cells in a fluorescence image slide using You Only Look Once version 5 (YOLOv5) with a feature pyramid network (FPN). Our proposed method can efficiently detect multiple cells with different sizes in a single image, eliminating the need for pixel-level segmentation. We show that our method outperforms state-of-the-art segmentation-based approaches in terms of accuracy and computational efficiency. The experimental results on publicly available datasets demonstrate that our proposed approach achieves an average precision of 0.8 and a processing time of 43.9 ms per image. Our approach addresses the research gap in the literature by providing a more efficient and accurate method for cell counting in fluorescence microscopy that requires less computational resources and labeled data.

show abstract

Attentional feature pyramid network for small object detection

Cited by 39 publications

References 61 publications

Is the diagnostic model based on convolutional neural network superior to pediatric radiologists in the ultrasonic diagnosis of biliary atresia?

Is the diagnostic model based on convolutional neural network superior to pediatric radiologists in the ultrasonic diagnosis of biliary atresia?

Mining the Micro-Trajectory of Two-Wheeled Non-Motorized Vehicles Based on the Improved YOLOx

YOLOv5-FPN: A Robust Framework for Multi-Sized Cell Counting in Fluorescence Images

Contact Info

Product

Resources

About