Deep Hashing Using Proxy Loss on Remote Sensing Image Retrieval

Xue, Shan; Li, Pingping; Wang, Yifan; Zhou, Qiuzhan; Wang, Zhen

doi:10.3390/rs13152924

Cited by 17 publications

(6 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…FMT-RAN [51], ST-RAN [51], FAH [52], DHPL [53], DHCNN [54], and AHCL [55] respectively. FMT-RAN trains the subsequent layers to be rotation invariant by feading the input image into the pretrained VGG and rotating the feature map generated by its last pooling layer generated by 4 different angles.…”

Section: ) Comparison Experiments With State-of-the-art Methodsmentioning

confidence: 99%

Proxy-Based Rotation Invariant Deep Metric Learning for Remote Sensing Image Retrieval

Cai,

Pan,

Jin

2024

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Convolutional Neural Networks (CNNs) are frequently utilized in content-based remote sensing image retrieval (CBRSIR). However, the features extracted by CNNs are not rotationally invariant, which is problematic for remote sensing (RS) images where objects appear at variable rotation angles. Additionally, because remote sensing images contain a wealth of content and detail information, CNNs may lead to information loss by superimposing multiple convolutional and pooling layers, affecting the ability of the model to extract features. To address these problems, this paper propose a proxybased feature fusion network (PBFFN). By designing a proxybased Euclidean distance contrast loss that combines contrast learning within the framework of metric learning, such that the distance between the source image and its rotated image embedding vector in the metric space is closer than any other image, thus endowing the model with a certain degree of rotation invariant. Meanwhile, the global correlation map is generated by multi-layer fusion, under whose guidance the features of each layer are fused to improve the feature extraction capability of the model and to reduce the loss in the image flow process. Extensive experiments based on two public remote sensing datasets show that the method achieves better performance compared to other methods.

show abstract

Section: ) Comparison Experiments With State-of-the-art Methodsmentioning

confidence: 99%

Proxy-Based Rotation Invariant Deep Metric Learning for Remote Sensing Image Retrieval

Cai,

Pan,

Jin

2024

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

show abstract

“…In contrast, supervised hashing methods often achieves better performance than unsupervised hashing methods. To address the problem that deep hashing networks tends to be highly expensive in terms of storage space and computing resources, Li et Shan et al have presented a proxy-based hash retrieval method, called DHPL (Deep Hashing using Proxy Loss), which combines hash code learning with proxy-based metric learning in a CNN [108]. Liu et al proposed a new RSIR method named feature and hash (FAH) learning, which consists of a deep feature learning model (DFLM) and an adversarial hash learning model (AHLM) [109].…”

Section: ) Hashing-based Methodsmentioning

confidence: 99%

Remote Sensing Image Retrieval in the Past Decade: Achievements, Challenges, and Future Directions

Zhou

Guan

et al. 2023

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Remote sensing image retrieval (RSIR) aims to search and retrieve the images of interest from a large remote sensing (RS) image archive, which has remained to be a hot topic over the past decade. Benefited from the advent and progress of deep learning, RSIR has been promoted by developing novel approaches, constructing new datasets, and exploring potential applications. To the best of our knowledge, there lacks a comprehensive review of RSIR achievements, including systematic and hierarchical categorization of RSIR methods and benchmark datasets in the past decade. This article therefore provides a systematic survey of the recently published RSIR methods and benchmarks by reviewing more than 200 papers. To be specific, in terms of image source, label, and modality, we first group the RSIR methods into some hierarchical categories, each of which is reviewed in detail. Following the categorization of the RSIR methods, we list the benchmark datasets publically available for performance evaluation, and present our newly collected RSIR dataset. Moreover, some of the existing RSIR methods are selected and evaluated on the representative benchmark datasets. The results demonstrate that deep learning-based methods are currently the dominant RSIR approaches and outperform handcrafted feature-based methods by a significant margin. Finally, we discuss the main challenges of RSIR, and point out some potential directions for the future RSIR research.

show abstract

“…To ensure the integrity of remote sensing images, we must use guided filtering to conduct linear processing of image data according to its overall characteristics [12] G as shown in (2) below.…”

Section: Processing Remote Sensing Image Target Data Of Power Towermentioning

confidence: 99%

Target feature matching method of remote sensing image of power tower supported by grid coding

Peng,

Chen,

et al. 2023

International Conference on Automation Control, Algorithm, and Intelligent Bionics (ACAIB 2023)

View full text Add to dashboard Cite

The conventional target feature matching method of remote sensing image of power tower mainly uses deformable pooling framework for sampling processing, which is vulnerable to the dynamic deformation of geographical space, resulting in low registration rate. Therefore, a new target feature matching method of remote sensing image of power tower needs to be designed with the support of grid coding, that is, combining the principle of thermal infrared image. The target data of remote sensing image of power tower is processed, and the target feature matching algorithm of remote sensing image of power tower is designed on the basis of grid coding support, so that the target feature matching of remote sensing image of power tower is realized. The experimental results show that the designed remote sensing image target feature matching method for power tower has a high registration rate, which proves that the designed remote sensing image target feature matching method has good matching effect, accuracy, and certain application value, and has made certain contributions to improving the quality of remote sensing images in various fields.

show abstract

Deep Hashing Using Proxy Loss on Remote Sensing Image Retrieval

Cited by 17 publications

References 36 publications

Proxy-Based Rotation Invariant Deep Metric Learning for Remote Sensing Image Retrieval

Proxy-Based Rotation Invariant Deep Metric Learning for Remote Sensing Image Retrieval

Remote Sensing Image Retrieval in the Past Decade: Achievements, Challenges, and Future Directions

Target feature matching method of remote sensing image of power tower supported by grid coding

Contact Info

Product

Resources

About