少即是多：用于DeepGlobe卫星土地覆盖分割的轻量级卷积神经网络的受控基准测试
When Less Is More: A Controlled Benchmark of Lightweight CNNs for Satellite Land-Cover Segmentation on DeepGlobe
浏览论文内容
中文总结 AI 辅助
研究在DeepGlobe数据集上比较VGG16等五种架构用于卫星土地覆盖分割，经三个迭代分离相关因素，采用相同预处理等协议。结果显示MobileNetV2_v1表现优，轻量级迁移学习模型在资源受限遥感环境中可匹配或超越更深模型，利于土地覆盖映射。
中文摘要 AI 辅助
高分辨率卫星图像是良好土地覆盖分类的支柱，没有它，环境监测、城市规划和可持续资源管理都会受到影响。深度学习架构在语义分割中表现良好，但在可控、可重复的条件下，经典卷积编码器的效率-准确性权衡尚未得到很好的量化。本研究在DeepGlobe土地覆盖分类数据集上比较了五种架构VGG16、MobileNetV2、InceptionV3、AlexNet和CNN，通过三个逐步优化的迭代来分离正则化、迁移学习和架构深度。为确保性能差异反映架构属性，所有实验使用相同的预处理、超参数和训练协议，不进行数据增强或类别不平衡校正。MobileNetV2_v1以24.98MB的模型大小获得了最高的总体准确率（0.7906）和平均交并比（0.4625），优于InceptionV3_v2（125.17MB，准确率0.7610）和VGG16_v2（71.13MB，准确率0.7653）等更深的架构。类别分析表明在城市、农业和水类别中有优势，但牧场- barren的混淆表明仅架构优化无法优化光谱相似的少数类别。在保留的测试图像上证实了强大的空间泛化和清晰的边界描绘，验证了操作适用性。这些结果表明，在资源受限的遥感环境中，轻量级、迁移学习的模型可以匹配或优于更深的模型，实现可扩展的土地覆盖映射。
英文摘要
High-resolution satellite imagery is the backbone of good land-cover classification, and without that, environmental monitoring, urban planning, and sustainable resource management all fall short. Deep learning architectures perform well in semantic segmentation, but the efficiency-accuracy trade-off across classical convolutional encoders is not well quantified under controlled, reproducible conditions. This study compares five architectures VGG16, MobileNetV2, InceptionV3, AlexNet, and CNN on the DeepGlobe Land Cover Classification dataset using three progressively optimized iterations to isolate regularisation, transfer learning, and architectural depth. To ensure performance differentials reflect architectural properties, all experiments used identical preprocessing, hyperparameter, and training protocols without data augmentation or class-imbalance correction. At 24.98 MB, MobileNetV2_v1 had the highest overall accuracy (0.7906) and mean Intersection over Union (0.4625), outperforming deeper alternatives like InceptionV3_v2 (125.17 MB, accuracy 0.7610) and VGG16_v2 (71.13 MB, accuracy 0.7653). Class-wise analysis showed strength in urban, agricultural, and water categories, but rangeland-barren confusion showed that architectural optimization alone cannot optimize spectrally similar minority classes. Strong spatial generalization and crisp boundary delineation were confirmed on held-out test imagery, validating operational applicability. These results show that lightweight, transfer-learned models can match or outperform deeper models in resource-constrained remote-sensing environments, enabling scalable land-cover mapping.
