With the development of deep learning, change detection technology has gained great progress. However, how to effectively extract multi-scale substantive changed features and accurately detect small objects as well accurate details is still a challenge. To solve problem, we propose Attentived Differential High-Resolution Change Detection Network (ADHR-CDNet) for remote sensing images. In ADHR-CDNet, novel high-resolution backbone with Pyramid Module (DPM) proposed multi-level features. The...

In this article, we present a new pansharpening method, zero-reference generative adversarial network (ZeRGAN), which fuses low spatial resolution multispectral (LR MS) and high panchromatic (PAN) images. the proposed indicates that it does not require paired reduced-scale images or unpaired full-scale for training. To obtain accurate fusion results, establish an game between set of multiscale generators their corresponding discriminators. Through generators, fused MS (HR are progressively...

With the emergence of large pre-trained vison-language model like CLIP, transferable representations can be adapted to a wide range downstream tasks via prompt tuning. Prompt tuning tries probe beneficial information for from general knowledge stored in model. A recently proposed method named Context Optimization (CoOp) introduces set learnable vectors as text language side. However, alone only adjust synthesized "classifier", while computed visual features image encoder not affected , thus...

Catastrophic forgetting is a critical chanllenge for incremental object detection (IOD). Most existing methods treat the detector monolithically, relying on instance replay or knowledge distillation without analyzing component-specific forgetting. Through dissection of Faster R-CNN, we reveal key insight: predominantly localized to RoI Head classifier, while regressors retain robustness across stages. This finding challenges conventional assumptions, motivating us develop framework termed...

A critical challenge for multi-modal Object Re-Identification (ReID) is the effective aggregation of complementary information to mitigate illumination issues. State-of-the-art methods typically employ complex and highly-coupled architectures, which unavoidably result in heavy computational costs. Moreover, significant distribution gap among different image spectra hinders joint representation multimodal features. In this paper, we propose a framework named as PromptMA establish...

In this paper, we construct a new coupled sparse non-negative matrix factorization (CSNMF) model for the fusion of panchromatic (PAN) and multispectral (MS) images. Two CSNMFs are developed joint representation MS PAN Moreover, sequential iterative algorithm is proposed to simultaneously find solution CSNMF. Because learned dictionaries can reveal latent structure images in spatial spectral domains, fused high-resolution be calculated by multiplying dictionary image coefficients Some...

Fusion of a panchromatic (PAN) image and corresponding multispectral (MS) is also known as pansharpening, which aims to combine abundant spatial details PAN spectral information MS images. Due the absence high-resolution images, available deep-learning-based methods usually follow paradigm training at reduced resolution testing both full resolution. When taking original images inputs, they always obtain sub-optimal results due scale variation. In this paper, we propose explore...

In this paper, a new pansharpening method is proposed by constructing set of multiscale geometric support tensor filters (MGSTFs). First, least-square ridgelet machine developed to derive series MGSTFs. Then the source images are formulated as tensors and filtered MGSTFs capture salient features images. These then fused at each scale direction obtain products. The distortions can be reduced exploring formulation multispectral data endowing filters' directionality details Some experiments...

The aim of multispectral (MS) and panchromatic (PAN) image fusion is to obtain an MS that has high resolution in both spectral spatial domains. During the process, there are two important issues, i.e., information preservation enhancement. In this article, we propose a dual-collaborative model considers not only correlation collaboration but also spatial-spectral collaboration. First, features PAN images extracted by shared feature embedding network. Then, order enhance details, decomposed...

Pansharpening is a feasible way to obtain the high-resolution (HR) multispectral (MS) images by using panchromatic (PAN) sharpen low-resolution MS images. Despite its great advances, most existing pansharpening methods neglect importance of integrating local and non-local characteristics images, resulting in imbalance spatial spectral distribution. In this paper, we propose complementary fusion network (CFNet) based on frequency hybrid attention mechanism for pansharpening. By introducing...

Download This Paper Open PDF in Browser Add to My Library Share: Permalink Using these links will ensure access this page indefinitely Copy URL DOI

The fusion of hyperspectral image (HSI) and multispectral (MSI) refers to enhance the spatial resolution HSI with help a corresponding MSI that has high finally obtain an in both spectral domains. In this letter, we propose variational tensor subspace decomposition-based method fully explore differences correlations among three modes tensor. Experimental results on two datasets show proposed can achieve superior performance compared existing state-of-the-art methods computational efficiency.

The process of fusing a high spatial resolution (HR) panchromatic (PAN) image and low (LR) multispectral (MS) to obtain an HRMS is known as pansharpening. With the development convolutional neural networks, performance pansharpening methods has been improved, however, blurry effects spectral distortion still exist in their fusion results due insufficiency details learning frequency mismatch between MS PAN. Therefore, improvement at premise reducing challenge. In this paper, we propose...

Image-based geo-localization is estimating the location of a query image by matching it to large amount images in geo-tagged database. This task very challenging due vast differences visual appearance or modality pairs on different platforms, for example, one from RGB camera, other light detection and ranging (LiDAR) sensor. The spatial layout scene can provide important clues significantly reduce ambiguity. Therefore, we propose novel deep network that embeds configuration scenes into...

As a novel succinate dehydrogenase inhibitor (SDHI), pydiflumetofen (PYD) exhibits broad-spectrum bactericidal activity in various crops; however, little is yet known about its absorption, translocation, and metabolic behavior within plants. Cucumber tomato plants were cultured hydroponic conditions spiked at 0.5 mg/L of PYD, samples collected certain intervals to investigate the residual fate PYD The results demonstrated that was readily absorbed by roots both plants, with mean root...