基于tensorflow深度学习VGG-19图像风格迁移+自动去噪(MNIST数据集)机器学习+人工智能+神经网络

上传者: 31136513 | 上传时间: 2025-05-19 13:15:43 | 文件大小: 522.16MB | 文件类型: ZIP
1.项目基于 MNIST 数据集,使用 VGG-19 网络模型,将图像进行风格迁移,实现去噪功能。 2.项目运行环境:Python 和 TensorFlow 运行环境。需要 Python 3.6 及以上配置,使用conda安装环境 conda create -n tensorflow python=3.8.10 3.项目包括 3 个模块:图片处理、模型构造、迭代更新。项目用到的网络模型为预训练好的VGG-19,使用过程中抛弃最后三个全连接层,取出前面各层的参数,构建网络结构。损失函数,由内容损失、风格损失构成。内容损失采用 L2范数损失,风格损失用 Gram 矩阵计算各通道的相关性,以便更好的捕捉笔触、纹理等细节信息,利用 adam 梯度下降算法进行优化。 4.准确率评估:对于图像风格迁移这种模糊算法,并没有客观的评判标准。损失函数可以反映出一部分情况,更多的是人为观察运行结果。经测试,经过 40 次迭代风格迁移已很明显,可根据自身需求,合理调节迭代次数。

文件下载

资源详情

[{"title":"( 56 个子文件 522.16MB ) 基于tensorflow深度学习VGG-19图像风格迁移+自动去噪(MNIST数据集)机器学习+人工智能+神经网络","children":[{"title":"基于tensorflow深度学习VGG-19图像风格迁移+自动去噪","children":[{"title":"style_transfer","children":[{"title":"utils.py <span style='color:#111;'> 2.22KB </span>","children":null,"spread":false},{"title":"style_transfer.py <span style='color:#111;'> 6.60KB </span>","children":null,"spread":false},{"title":"自编码器去噪.ipynb <span style='color:#111;'> 89.19KB </span>","children":null,"spread":false},{"title":".idea","children":[{"title":"web_IST.iml <span style='color:#111;'> 556B </span>","children":null,"spread":false},{"title":"workspace.xml <span style='color:#111;'> 9.54KB </span>","children":null,"spread":false},{"title":"misc.xml <span style='color:#111;'> 292B </span>","children":null,"spread":false},{"title":"modules.xml <span style='color:#111;'> 266B </span>","children":null,"spread":false}],"spread":true},{"title":"imagenet-vgg-verydeep-19.mat <span style='color:#111;'> 510.12MB </span>","children":null,"spread":false},{"title":"load_vgg.py <span style='color:#111;'> 3.51KB </span>","children":null,"spread":false},{"title":"__pycache__","children":[{"title":"utils.cpython-35.pyc <span style='color:#111;'> 3.29KB </span>","children":null,"spread":false},{"title":"style_transfer.cpython-35.pyc <span style='color:#111;'> 7.65KB </span>","children":null,"spread":false},{"title":"load_vgg.cpython-36.pyc <span style='color:#111;'> 2.86KB </span>","children":null,"spread":false},{"title":"load_vgg.cpython-35.pyc <span style='color:#111;'> 3.84KB </span>","children":null,"spread":false},{"title":"style_transfer.cpython-36.pyc <span style='color:#111;'> 6.40KB </span>","children":null,"spread":false},{"title":"utils.cpython-36.pyc <span style='color:#111;'> 2.50KB </span>","children":null,"spread":false}],"spread":true},{"title":"static","children":[{"title":"gif","children":[{"title":"architechture_pattern.gif <span style='color:#111;'> 2.69MB </span>","children":null,"spread":false}],"spread":true},{"title":"styles","children":[{"title":"resized_pattern.jpg <span style='color:#111;'> 55.19KB </span>","children":null,"spread":false},{"title":"harlequin.jpg <span style='color:#111;'> 140.73KB </span>","children":null,"spread":false},{"title":"resized_scream.jpg <span style='color:#111;'> 26.90KB </span>","children":null,"spread":false},{"title":"resized_starry_night.jpg <span style='color:#111;'> 33.62KB </span>","children":null,"spread":false},{"title":"dusk.jpg <span style='color:#111;'> 119.61KB </span>","children":null,"spread":false},{"title":"pattern.jpg <span style='color:#111;'> 502.71KB </span>","children":null,"spread":false},{"title":"resized_dusk.jpg <span style='color:#111;'> 15.25KB </span>","children":null,"spread":false},{"title":"guernica.jpg <span style='color:#111;'> 98.78KB </span>","children":null,"spread":false},{"title":"scream.jpg <span style='color:#111;'> 743.55KB </span>","children":null,"spread":false},{"title":"starry_night.jpg <span style='color:#111;'> 598.96KB </span>","children":null,"spread":false}],"spread":true},{"title":"contents","children":[{"title":"architechture.jpg <span style='color:#111;'> 35.25KB </span>","children":null,"spread":false},{"title":"resized_sky.jpg <span style='color:#111;'> 20.18KB </span>","children":null,"spread":false},{"title":"resized_scenery.jpg <span style='color:#111;'> 232.69KB </span>","children":null,"spread":false},{"title":"scenery.jpg <span style='color:#111;'> 118.01KB </span>","children":null,"spread":false},{"title":"marvel.jpg <span style='color:#111;'> 266.09KB </span>","children":null,"spread":false},{"title":"resized_architechture.jpg <span style='color:#111;'> 26.88KB </span>","children":null,"spread":false},{"title":"sky.jpg <span style='color:#111;'> 87.14KB </span>","children":null,"spread":false},{"title":"resized_yuyuantan.jpg <span style='color:#111;'> 16.51KB </span>","children":null,"spread":false},{"title":"deadpool.jpg <span style='color:#111;'> 104.51KB </span>","children":null,"spread":false}],"spread":true},{"title":"outputs","children":[{"title":"architechture_pattern","children":[{"title":"epoch_280.png <span style='color:#111;'> 310.39KB </span>","children":null,"spread":false},{"title":"epoch_5.png <span style='color:#111;'> 259.04KB </span>","children":null,"spread":false},{"title":"epoch_240.png <span style='color:#111;'> 309.93KB </span>","children":null,"spread":false},{"title":"epoch_60.png <span style='color:#111;'> 303.21KB </span>","children":null,"spread":false},{"title":"epoch_20.png <span style='color:#111;'> 284.97KB </span>","children":null,"spread":false},{"title":"epoch_260.png <span style='color:#111;'> 310.18KB </span>","children":null,"spread":false},{"title":"epoch_3.png <span style='color:#111;'> 255.57KB </span>","children":null,"spread":false},{"title":"epoch_160.png <span style='color:#111;'> 308.45KB </span>","children":null,"spread":false},{"title":"epoch_100.png <span style='color:#111;'> 306.38KB </span>","children":null,"spread":false},{"title":"epoch_10.png <span style='color:#111;'> 268.11KB </span>","children":null,"spread":false},{"title":"epoch_180.png <span style='color:#111;'> 308.90KB </span>","children":null,"spread":false},{"title":"epoch_300.png <span style='color:#111;'> 310.60KB </span>","children":null,"spread":false},{"title":"epoch_2.png <span style='color:#111;'> 254.00KB </span>","children":null,"spread":false},{"title":"epoch_1.png <span style='color:#111;'> 252.54KB </span>","children":null,"spread":false},{"title":"epoch_4.png <span style='color:#111;'> 257.26KB </span>","children":null,"spread":false},{"title":"epoch_140.png <span style='color:#111;'> 307.93KB </span>","children":null,"spread":false},{"title":"epoch_120.png <span style='color:#111;'> 307.27KB </span>","children":null,"spread":false},{"title":"epoch_80.png <span style='color:#111;'> 305.21KB </span>","children":null,"spread":false},{"title":"epoch_200.png <span style='color:#111;'> 309.31KB </span>","children":null,"spread":false},{"title":"epoch_40.png <span style='color:#111;'> 299.06KB </span>","children":null,"spread":false},{"title":"epoch_220.png <span style='color:#111;'> 309.66KB </span>","children":null,"spread":false}],"spread":false}],"spread":true}],"spread":true}],"spread":true}],"spread":true}],"spread":true}]

评论信息

免责申明

【只为小站】的资源来自网友分享,仅供学习研究,请务必在下载后24小时内给予删除,不得用于其他任何用途,否则后果自负。基于互联网的特殊性,【只为小站】 无法对用户传输的作品、信息、内容的权属或合法性、合规性、真实性、科学性、完整权、有效性等进行实质审查;无论 【只为小站】 经营者是否已进行审查,用户均应自行承担因其传输的作品、信息、内容而可能或已经产生的侵权或权属纠纷等法律责任。
本站所有资源不代表本站的观点或立场,基于网友分享,根据中国法律《信息网络传播权保护条例》第二十二条之规定,若资源存在侵权或相关问题请联系本站客服人员,zhiweidada#qq.com,请把#换成@,本站将给予最大的支持与配合,做到及时反馈和处理。关于更多版权及免责申明参见 版权及免责申明