Detection of Objects from Noisy Images

Nayan, Al-Akhir; Saha, Joyeta; Mahmud, Khan Raqib; Azad, Abul K.; Kibria, Muhammad Golam

doi:10.1109/sti50764.2020.9350521

Cited by 8 publications

(5 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…"Detection of Objects from Noisy Images [7]" proposes an object detection algorithm for noisy images. It describes the low cost and efficient method for finding objects in the noisy image.…”

Section: Literature Reviewmentioning

confidence: 99%

WITHDRAWN: Object Extraction and Detection UsingU2-Net and YOLOv7

Bhatt,

Makwana,

Joshi

et al. 2024

Preprint

View full text Add to dashboard Cite

Object extraction, detection and mattingof background are constitute foundational pursuits incomputer vision across diverse application domains.For serving this purpose we delves into the usageof U2-Net, a deep learning model, and YOLOv7, anadvanced object detection framework. This attempt ofinvestigation of YOLOv7, an advanced object recogni-tion framework, and U2-Net, a deep learning networkwith a focus on noticeable object extraction. In ourstudy, U2-Net and YOLOv7 are trained on mentioneddata-sets, with a focus on their unique contributionsto object extraction and detection and matting. Theprimary focus of this study is the practical exami-nation of their application, with particular attentionto their real-world implementation along with associ-ated challenges. Notably, we emphasize our intentionto reinforce this effort by providing these modelswith domain-specific information through training onpertinent data-sets. We carefully examine how U2-Net and YOLOv7 could improve the object detectionand recognition performance of our study. The mainobjective of this research is to boost imagination inapplications involving computer vision by providingreal-world examples of how these models might beused.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

WITHDRAWN: Object Extraction and Detection UsingU2-Net and YOLOv7

Bhatt,

Makwana,

Joshi

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…However, relying solely on data-level enhancements has limitations in significantly improving small object detection accuracy and may introduce unwanted noise. Another approach, known as multi-scale learning, combines spatial details in the shallow layers and semantic details in the deep layers to tackle the imbalance problem at the feature level [20][21][22]. The objective of this approach is to maximize the utilization of diverse scales, harnessing their complementary nature, with the ultimate goal of enhancing the detection ability of small objects.…”

Section: Small Object Detectionmentioning

confidence: 99%

Composite Backbone Small Object Detection Based on Context and Multi-Scale Information with Attention Mechanism

Jing,

Liu,

Liu

2024

Mathematics

View full text Add to dashboard Cite

Object detection has gained widespread application across various domains; nevertheless, small object detection still presents numerous challenges due to the inherent limitations of small objects, such as their limited resolution and susceptibility to interference from neighboring elements. To improve detection accuracy of small objects, this study presents a novel method that integrates context information, attention mechanism, and multi-scale information. First, to realize feature augmentation, a composite backbone network is employed which can jointly extract object features. On this basis, to efficiently incorporate context information and focus on key features, the composite dilated convolution and attention module (CDAM) is designed, consisting of a composite dilated convolution module (CDM) and convolutional block attention module (CBAM). Then, a feature elimination module (FEM) is introduced to reduce the feature proportion of medium and large objects on feature layers; the impact of neighboring objects on small object detection can thereby be mitigated. Experiments conducted on MS COCO validate the superior performance of the method compared with baseline detectors, while it yields an average enhancement of 0.8% in overall detection accuracy, with a notable enhancement of 2.7% in small object detection.

show abstract

“…For instance, megacosm is a set of annotated pictures with the positions of various celestial bodies-but it contains insufficient images (400) [12]. • Astronomical images are noisy, and methods like YOLO are sensitive to noise and need to be trained on realistic datasets [13]. To build an effective training set, using high-quality images such as those obtained with Hubble and/or the James Webb Space Telescope is not relevant, and adding artificial noise to these images is not effective either because it will not be as realistic as real noise.…”

Section: Related Workmentioning

confidence: 99%

DeepSpaceYoloDataset: Annotated Astronomical Images Captured with Smart Telescopes

Parisot

2024

Data

View full text Add to dashboard Cite

Recent smart telescopes allow the automatic collection of a large quantity of data for specific portions of the night sky—with the goal of capturing images of deep sky objects (nebula, galaxies, globular clusters). Nevertheless, human verification is still required afterwards to check whether celestial targets are effectively visible in the images produced by these instruments. Depending on the magnitude of deep sky objects, the observation conditions and the cumulative time of data acquisition, it is possible that only stars are present in the images. In addition, unfavorable external conditions (light pollution, bright moon, etc.) can make capture difficult. In this paper, we describe DeepSpaceYoloDataset, a set of 4696 RGB astronomical images captured by two smart telescopes and annotated with the positions of deep sky objects that are effectively in the images. This dataset can be used to train detection models on this type of image, enabling the better control of the duration of capture sessions, but also to detect unexpected celestial events such as supernova.

show abstract

Detection of Objects from Noisy Images

Cited by 8 publications

References 17 publications

WITHDRAWN: Object Extraction and Detection UsingU2-Net and YOLOv7

WITHDRAWN: Object Extraction and Detection UsingU2-Net and YOLOv7

Composite Backbone Small Object Detection Based on Context and Multi-Scale Information with Attention Mechanism

DeepSpaceYoloDataset: Annotated Astronomical Images Captured with Smart Telescopes

Contact Info

Product

Resources

About