multispectral data augmentation

本文探讨了通过合成数据与少量真实红外数据结合的方法来解决热成像中无人机检测难题，强调了数据集对齐的重要性。

This work addresses the limited generalization of existing deep models in multispectral (visible and thermal infrared) video surveillance, which stems from sensor discrepancies and the scarcity of thermal imaging data. The authors propose a CNN-based framework for multispectral object detection and systematically design and evaluate several cross-spectral data augmentation strategies—namely thermal feature simulation, texture-preserving transformations, and illumination-invariant enhancement—to effectively integrate color, shape, and thermal radiation cues. Experimental results demonstrate that the proposed approach significantly improves detection accuracy and robustness in mixed-spectral scenarios, validates the auxiliary value of visible-spectrum data for thermal infrared detection, and fills a critical gap in the understanding of cross-spectral data augmentation mechanisms.

The scarcity of real-world thermal imagery—due to high acquisition costs and limited scene coverage—severely constrains deep learning advancement in drone-based thermal imaging. Method: This paper introduces the first procedural thermal image synthesis pipeline tailored for aerial perspectives, integrating 3D pose alignment with physics-based thermal radiation modeling to embed arbitrary target classes (e.g., drones, wildlife) into authentic thermal backgrounds with precise control over position, scale, and viewpoint. The approach ensures class extensibility, geometric fidelity, and physically plausible thermal characteristics. Results: Augmenting the HIT-UAV and MONET datasets with synthetically injected novel categories significantly improves object detection performance; models trained exclusively on synthetic thermal data outperform visible-light baselines, demonstrating the efficacy and generalizability of the synthesized data in challenging conditions such as low illumination and occlusion.

研究通过使用生成模型（如GANs、ControlNet扩散模型等）将RGB图像转换为IR图像，以增强红外车辆检测在未知UAV领域的性能，有效缓解了IR数据稀缺问题。

To address three key limitations in generative image forgery detection—weak artifact features, model overfitting, and insufficient local awareness—this paper proposes SAFE, a lightweight and efficient detection framework. Methodologically, SAFE redefines the training paradigm of Source Identification (SID) from an image transformation perspective: it replaces conventional downsampling with cropping for preprocessing, introduces enhanced data augmentations (ColorJitter and Rotation), and designs a patch-level random masking strategy tailored for SID; additionally, it employs a lightweight CNN architecture. Evaluated on an open-world benchmark covering 26 generative models, SAFE achieves state-of-the-art performance: +4.5% accuracy and +2.9% mean average precision over prior methods, demonstrating significantly improved robustness in detecting synthetic images.

本文通过文献综述和大规模实证研究，评估了50种图像增强技术作为深度学习图像检索系统的测试生成方法，以提高系统可靠性。

This work addresses the challenge of directly applying RGB-pretrained vision-language models to thermal infrared drone imagery by proposing a lightweight multimodal adaptation framework. The approach leverages multimodal projection alignment to transfer InternVL3 and Qwen-VL family models into the thermal infrared domain, followed by fine-tuning on real-world drone data to enable species identification, individual counting, and habitat semantic understanding. To the best of our knowledge, this is the first study to introduce lightweight projection-based adaptation for thermal infrared ecological monitoring, supporting high-accuracy inference under both closed-set and open-set prompting. Experiments show that Qwen3-VL-8B-Instruct achieves the best performance in open-set settings, yielding F1 scores of 0.968, 0.915, and 0.935 for elephants, rhinos, and deer, respectively, with perfect individual counting accuracy (1.000) and effective generation of contextual habitat descriptions.

This work addresses the lack of fine-grained diagnostic tools for evaluating the performance limitations of existing aerial object detectors in complex scenes. It introduces, for the first time, large-scale text-to-image generative models into the diagnostic pipeline of aerial detection systems, establishing a controllable synthetic testing platform. Through text-guided image generation, attribute-controllable editing, and automated validation, the framework enables systematic evaluation of pretrained vehicle detectors. The approach accurately predicts real-world performance deficiencies and effectively guides targeted data collection: augmenting the training set with only a small amount of carefully selected real data improves AP50 by up to 13%, substantially outperforming non-directed augmentation strategies. The modular and extensible design of the framework establishes a new paradigm for robustness analysis in aerial vision systems.

To address the insufficient robustness of deep vision models under common image corruptions, this paper proposes a data augmentation pipeline integrating neural style transfer with controllable synthetic image generation. We first observe that stylized degradation—though increasing Fréchet Inception Distance (FID)—significantly improves corruption robustness. We further uncover the complementary mechanisms between style transfer and synthetic data augmentation, and formally characterize their compatibility boundary with rule-based methods such as TrivialAugment. Through systematic hyperparameter analysis and cross-benchmark evaluation, our method achieves state-of-the-art robust accuracy on CIFAR-10-C (93.54%), CIFAR-100-C (74.90%), and TinyImageNet-C (50.86%), establishing new SOTA results on small-scale corruption benchmarks.

This work proposes a scalable evaluation framework to effectively assess the perceptual realism of synthetic images generated under adverse environmental conditions such as fog, rain, snow, and nighttime. The framework innovatively integrates a visual-language model (VLM) jury for perceptual realism scoring with distributional similarity analysis in embedding space, enabling, for the first time, a unified and efficient evaluation of both generative and rule-based image enhancement methods, with real images serving as the performance upper bound. Experimental results demonstrate that generative AI approaches significantly outperform rule-based methods, with the best-performing model achieving an acceptance rate approximately 3.6 times higher than that of rule-based counterparts; under most conditions, its synthetic images attain or even surpass the realism of real images.