Probabilistic faster R-CNN with stochastic region proposing: Towards object detection and recognition in remote sensing imagery

Yi, Dewei; Su, Jinya; Chen, Wenhua

doi:10.1016/j.neucom.2021.06.072

Cited by 36 publications

(14 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Chai focused on cloud and cloud shadow detection in Landsat data and proposed an approach based on deep CNNs (Chai et al, 2019 ). Han employed CNN in NWPU VHR-10 satellite data for target recognition and achieve great performance (Han et al, 2020 ), and Yi proposed a novel approach based on probabilistic faster R-CNN for object detection (Yi et al, 2021 ). Chen presented DRSNet (Chen and Tsou, 2021 ), an architecture for image scene classification in low-resolution data, and Swain implemented dimensionality reduction techniques to hyperspectral data (Swain and Banerjee, 2021 ).…”

Section: Related Studiesmentioning

confidence: 99%

Applying Digital Twins to Research the Relationship Between Urban Expansion and Vegetation Coverage: A Case Study of Natural Preserve

Zhao

Wang

et al. 2022

Front. Plant Sci.

View full text Add to dashboard Cite

With the growth of the world population, cities expand and encroach on forests and plants, causing many environmental problems. Digital Twin, as the rapidly developing technique in recent years, provides the opportunity to implement the specific situation of forests and plants at present or in the future, which has great performance on predictive analysis and optimization. From the consideration of plants and forests, this study provides a comprehensive case study to research the relationship between urban development boundary and natural environment in a natural preserve in a coastal city. Multispectral data of the study area is collected by Unmanned Aerial Vehicle (UAV), combining satellite remote sensing (RS) historical data and geographic data to establish the digital twin model for plant identification. In conjunction with local Master planning of land use, the results of modeling are used to analyze the influences of urban construction on the natural environment, and the inappropriate aspects of the planning are discovered and summarized. In addition, 6 suggestions for effective management and planning strategies are presented. As plants and forests are effective factors of natural conditions, this study offered an objective assessment for the sustainability and rationality of urban planning with some guidance and bases.

show abstract

Section: Related Studiesmentioning

confidence: 99%

Applying Digital Twins to Research the Relationship Between Urban Expansion and Vegetation Coverage: A Case Study of Natural Preserve

Zhao

Wang

et al. 2022

Front. Plant Sci.

View full text Add to dashboard Cite

show abstract

“…Other systems are affected by factors that cannot be determined a priori, or are liable to changes over time.Machine learning models therefore leverage knowledge represented in probabilistic forms to improve performance in these situations. For instance, in their Probabilistic faster R-CNN, Yi et al [24] incorporate a probabilistic region proposal network to stochastically predict the objectness of candidate windows in a Faster R-CNN [25] framework used for object detection from remote imagery. The probabilistic model used is Bayesian inference, which assigns confidence scores describing the uncertainty associated with each region proposal, allowing relevant more useful regions to be selected based on their quality, rather than the fixed threshold selection approach that characterizes conventional Faster R-CNN method.…”

Section: Knowledge Representationmentioning

confidence: 99%

Robust appearance modeling for object detection and tracking: a survey of deep learning approaches

Mumuni

2022

Prog Artif Intell

View full text Add to dashboard Cite

We review current and emerging knowledge-informed and brain-inspired cognitive systems for realizing adversarial defenses, eXplainable Artificial Intelligence (XAI), and zero-shot or few-shot learning. Data-driven deep learning models have achieved remarkable performance and demonstrated capabilities surpassing human experts in many applications. Yet, their inability to exploit domain knowledge leads to serious performance limitations in practical applications. In particular, deep learning systems are exposed to adversarial attacks, which can trick them into making glaringly incorrect decisions. Moreover, complex data-driven models typically lack interpretability or explainability, i.e., their decisions cannot be understood by human subjects. Furthermore, models are usually trained on standard datasets with a closed-world assumption. Hence, they struggle to generalize to unseen cases during inference in practical open-world environments, thus, raising the zero-or few-shot generalization problem. Although many conventional solutions exist, explicit domain knowledge, brain-inspired neural network and cognitive architectures offer powerful new dimensions towards alleviating these problems. Prior knowledge is represented in appropriate forms and incorporated in deep learning frameworks to improve performance. Brain-inspired cognition methods use computational models that mimic the human mind to enhance intelligent behavior in artificial agents and autonomous robots. Ultimately, these models achieve better explainability, higher adversarial robustness and data-efficient learning, and can, in turn, provide insights for cognitive science and neurosciencethat is, to deepen human understanding on how the brain works in general, and how it handles these problems.Index Terms-Domain knowledge, cognitive architecture, brain-inspired neural network, explainable AI, adversarial attack, zero-shot generalization. ✦ INTRODUCTION State of deep learningMachine learning is an artificial intelligence (AI) concept that enables computing systems to learn useful relationships from data and then use this information to identify learned patterns and make predictions on new inputs. Deep learning (DL) is a machine learning method that uses a multi-layered arrangement of computational units to learn relevant patterns from data. Many deep learning constructs have been proposed for various use-cases. Some of the most popular ones are deep convolutional neural networks (DCNN) [1], [2] recurrent neural networks (RNN) [3], generative adversarial networks (GANs) [4], deep reinforcement learning [5] and vision transformers [6], [7]). DCNNs are a class of deep neural networks (DNNs) specially designed to handle image data.Recently, deep learning methods have accomplished remarkable milestones in many critical applications. In medical domains, some state-of-the-art models (e.g., as reported

show abstract

“…Fast R-CNN extracts the features from the proposed regions and generates a class label and bounding box. Yi et al presented a probabilistic faster R-CNN technique with a stochastic region to recognise and locate grasshoppers from a remote sensing image and achieved a 0.9263 f1-score [11].…”

Section: Faster R-cnnmentioning

confidence: 99%

“…The Mask R-CNN [11] adapts a two-stage procedure, wherein the first stage is the RPN. In the next stage, in parallel to predicting the box offset and the class, Mask R-CNN also outputs a binary mask for each ROI.…”

Section: Mask R-cnnmentioning

confidence: 99%

An Efficient Deep Convolutional Neural Network Approach for Object Detection and Recognition Using a Multi-Scale Anchor Box in Real-Time

2021

View full text Add to dashboard Cite

Deep learning is a relatively new branch of machine learning in which computers are taught to recognize patterns in massive volumes of data. It primarily describes learning at various levels of representation, which aids in understanding data that includes text, voice, and visuals. Convolutional neural networks have been used to solve challenges in computer vision, including object identification, image classification, semantic segmentation and a lot more. Object detection in videos involves confirming the presence of the object in the image or video and then locating it accurately for recognition. In the video, modelling techniques suffer from high computation and memory costs, which may decrease performance measures such as accuracy and efficiency to identify the object accurately in real-time. The current object detection technique based on a deep convolution neural network requires executing multilevel convolution and pooling operations on the entire image to extract deep semantic properties from it. For large objects, detection models can provide superior results; however, those models fail to detect the varying size of the objects that have low resolution and are greatly influenced by noise because the features after the repeated convolution operations of existing models do not fully represent the essential characteristics of the objects in real-time. With the help of a multi-scale anchor box, the proposed approach reported in this paper enhances the detection accuracy by extracting features at multiple convolution levels of the object. The major contribution of this paper is to design a model to understand better the parameters and the hyper-parameters which affect the detection and the recognition of objects of varying sizes and shapes, and to achieve real-time object detection and recognition speeds by improving accuracy. The proposed model has achieved 84.49 mAP on the test set of the Pascal VOC-2007 dataset at 11 FPS, which is comparatively better than other real-time object detection models.

show abstract

Probabilistic faster R-CNN with stochastic region proposing: Towards object detection and recognition in remote sensing imagery

Cited by 36 publications

References 45 publications

Applying Digital Twins to Research the Relationship Between Urban Expansion and Vegetation Coverage: A Case Study of Natural Preserve

Applying Digital Twins to Research the Relationship Between Urban Expansion and Vegetation Coverage: A Case Study of Natural Preserve

Robust appearance modeling for object detection and tracking: a survey of deep learning approaches

An Efficient Deep Convolutional Neural Network Approach for Object Detection and Recognition Using a Multi-Scale Anchor Box in Real-Time

Contact Info

Product

Resources

About