Deep learning-based detection from the perspective of small or tiny objects: A survey

Tong, Kang; Wu, Yiquan

doi:10.1016/j.imavis.2022.104471

Cited by 101 publications

(30 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Object detection and recognition is one of the hot topics in computer vision [48]. Many challenges can be found for detecting the objects in an images including the scale of the objects, the similarity between some objects, and thee overlapping between them [49]. Object detection models that are optimal for detection need to have a higher input network size for smaller object.…”

Section: Object Detectionmentioning

confidence: 99%

“…In the literature, there is luck of papers that compared the proposed features extraction networks for deep-learningbased techniques [5,10] . For computer vision tasks , the choose of suitable network (Backbone) for features extraction can be costly, due to the fact that some tasks are used some specific backbones while its not suitable for others [48,49]. In this paper, we attempted to collect and describe various existing backbones used for features extraction.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Backbones-Review: Feature Extraction Networks for Deep Learning and Deep Reinforcement Learning Approaches

Elharrouss¹,

Akbari²,

Almaadeed³

et al. 2022

Preprint

View full text Add to dashboard Cite

To understand the real world using various types of data, Artificial Intelligence (AI) is the most used technique nowadays. While finding the pattern within the analyzed data represents the main task. This is performed by extracting representative features step, which is proceeded using the statistical algorithms or using some specific filters. However, the selection of useful features from large-scale data represented a crucial challenge. Now, with the development of convolution neural networks (CNNs), the feature extraction operation has become more automatic and easier. CNNs allow to work on large-scale size of data, as well as cover different scenarios for a specific task. For computer vision tasks, convolutional networks are used to extract features also for the other parts of a deep learning model. The selection of a suitable network for feature extraction or the other parts of a DL model is not random work. So, the implementation of such a model can be related to the target task as well as the computational complexity of it. Many networks have been proposed and become the famous networks used for any DL models in any AI task. These networks are exploited for feature extraction or at the beginning of any DL model which is named backbones. A backbone is a known network trained in many other tasks before and demonstrates its effectiveness. In this paper, an overview of the existing backbones, e.g. VGGs, ResNets, DenseNet, etc, is given with a detailed description. Also, a couple of computer vision tasks are discussed by providing a review of each task regarding the backbones used. In addition, a comparison in terms of performance is also provided, based on the backbone used for each task.

show abstract

Section: Object Detectionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Backbones-Review: Feature Extraction Networks for Deep Learning and Deep Reinforcement Learning Approaches

Elharrouss¹,

Akbari²,

Almaadeed³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…In most detection scenarios, the detection of small objects accounts for the majority. There are many reasons why small objects are difficult to detect [42]. First, the small objects occupy a small area of the digital image, and the feature information is insufficient [43].…”

Section: Related Workmentioning

confidence: 99%

“…Second, small objects and large and medium-sized objects exist in the same detection scene at the same time, and the size span between targets is large. Finally, there is a large gap between the scale number of small objects available for training and the small object classification [42]. Many methods have been proposed to solve the problems faced by small objects [42].…”

Section: Related Workmentioning

confidence: 99%

“…Finally, there is a large gap between the scale number of small objects available for training and the small object classification [42]. Many methods have been proposed to solve the problems faced by small objects [42]. The common solutions are as follows: changing the training strategy and carrying out multi-scale training to solve the problem of object scale span.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Multiple spatial residual network for object detection

Dong

Jiang

Tao

et al. 2022

Complex Intell. Syst.

View full text Add to dashboard Cite

Many residual network-based methods have been proposed to perform object detection. However, most of them may lead to overfitting or cannot perform well in small object detection and alleviate the problem of overfitting. We propose a multiple spatial residual network (MSRNet) for object detection. Particularly, our method is based on central point detection algorithm. Our proposed MSRNet employs a residual network as the backbone. The resulting features are processed by our proposed residual channel pooling module. We then construct a multi-scale feature transposed residual fusion structure consists of three overlapping stacked residual convolution modules and a transpose convolution function. Finally, we use the Center structure to process the high-resolution feature image for obtaining the final prediction detection result. Experimental results on PASCAL VOC dataset and COCO dataset confirm that the MSRNet has competitive accuracy compared with several other classical object detection algorithms, while providing a unified framework for training and reasoning. The MSRNet runs on GeForce RTX 2080Ti.

show abstract

YOLOv7‐EAS: A Small Target Detection of Camera Module Surface Based on Improved YOLOv7

Zou

Yuan

et al. 2023

Advcd Theory and Sims

View full text Add to dashboard Cite

The camera module surface may introduce defects in the focusing process, and manual detection of these defects is inefficient and costly. To address this problem, a method to automatically expand the dataset for deep learning models with fewer original images is developed. A defect detection method is also presented, YOLOv7‐EAS, that enhances YOLOv7 for camera module images with complex backgrounds and small target defects. This method introduces NewEIOU Loss, a bounding box regression loss function that extends EIOU Loss with a nonlinear compensation term. This function improves the model's convergence rate to quadratic order. Moreover, the detection head based on ASFF is redesigned to adapt the layer features and spatially filter out irrelevant information to improve the detection rate of small targets. Finally, a 3D attention mechanism is applied to increase the model's ability to extract fine‐grained features without increasing the network parameters. Experimental results show that the improved YOLOv7 algorithm achieves 90.5% mAP for the camera module image dataset, which is 10.7% higher than the original algorithm, at 136 FPS. This method can perform the camera module surface defect detection task more accurately and meet the real‐time requirements of the production line.

show abstract

Deep learning-based detection from the perspective of small or tiny objects: A survey

Cited by 101 publications

References 53 publications

Backbones-Review: Feature Extraction Networks for Deep Learning and Deep Reinforcement Learning Approaches

Backbones-Review: Feature Extraction Networks for Deep Learning and Deep Reinforcement Learning Approaches

Multiple spatial residual network for object detection

YOLOv7‐EAS: A Small Target Detection of Camera Module Surface Based on Improved YOLOv7

Contact Info

Product

Resources

About