Neural Style Transfer (NST) and Steganography Dataset for NCSM

This dataset consists of 17 content images and 16 style images created using a combination of DALL·E-generated AI artwork and hand-drawn elements to ensure originality. The images were digitally processed and formatted to be compatible with Neural Style Transfer (NST) models. The dataset is designed for research on style transfer and steganography, specifically within the Neural Crypto Stego Model (NCSM). In this study, an encrypted message was embedded into a transformed content image using Least Significant Bit (LSB) steganography. The dataset provides a valuable resource for evaluating and comparing different NST methodologies and their effectiveness in secure communication.本数据集包含17张内容图像与16张风格图像，通过结合DALL·E生成的AI艺术作品与手绘元素制作而成，以保障原创性。所有图像均经过数字化处理与格式适配，可兼容神经风格迁移（Neural Style Transfer, NST）模型。 本数据集专为风格迁移与隐写术研究设计，尤其适用于神经加密隐写模型（Neural Crypto Stego Model, NCSM）相关研究。在此项研究中，研究人员通过最低有效位（Least Significant Bit, LSB）隐写术，将加密消息嵌入至经过变换的内容图像中。本数据集为评估与对比各类神经风格迁移方法，及其在安全通信中的有效性提供了宝贵的研究资源。