资源列表
53607888mnistclassify_DBN
- 目前深度学习已经得到研究者的广泛关注，在这里深信度网络被用于手写字符的识别，得到非常好的分类精度。-Currently deep learning has been much attention, here we are convinced of the network is used to identify handwritten characters, get very good classification accuracy.
CNN-pooling-strategy
- 基于卷积层和池化层的卷积深度网络被执行，该框架可以有效地识别灰度图像，彩色图像和高光谱图像。- Convolution deep network based on convolution layer and pooling layer is performed, the framework can effectively identify grayscale images, color images and hyperspectral images.
SAE_DBN_CNNToolbox
- 多种深度学习框架，主要包括堆栈稀疏自动编码器，深信度网络，卷积神经网络等。对于灰度图像和高维图像，展现非常强大的学习性能。-A variety of deep learning framework, including automatic stack sparse encoder, is convinced of the network, convolution neural networks. For grayscale images and high-dimensional image, s
mvcnn-master
- 3DShapeNetsCode neuronal network for 3D mesh retr
huahengjiance
- 在复杂背景下的多种类疵病检测并标出疵病数目-Many types of flaw detection in complex background and mark the number of flaws
20132085
- 利用D3D编程的一个简单的3D小球升降程序，包含了上下的控制，新手可以-Using D3D programming a simple 3 d ball lifting procedure, contains the up and down control, beginners can take a look
coders
- 使用D3D编写的简单开车游戏，新手可以看看，做的比较简陋，但应有的功能还是有的，包含开始界面，赛车选择，有背景音乐，可切换视角-Using D3D write simple driving game, beginners can see, the more humble, but its function or some, contains the start interface, the car selection, background music, can switch perspecti
20132085_test6
- 使用MFC AppWizard 建立一个SDI 程序，。可增加绘制图形球体或茶壶等，可在环境中创建点光源、平行光、聚光灯（可利用对话框输入参数创建）、设置所绘制对象的材质，呈现出塑料、金属等材质特性。可对光源的位置、方向、类型进行控制，改变材质参数-A SDI procedures, using MFC AppWizard to create. Can increase drawing graphics sphere or teapot, etc, can create a point sour
20132085
- 使用Visual C++建立一个单文档程序，程序中有图形变换菜单（各基本变换对应菜单项），各图形变换的参数通过对话框提供（设计提供变换参数的对话框，如平移、旋转、比例等），实现二维图形的变换。-Program using Visual c++ to create a single document, in the program have a menu graphics transform (the basic transform corresponding menu items), graph
Thimelg
- This study presents a geometric model for segmentation of ultrasound images. A partial differential equation based flow is designed in order to achieve a maximum likelihood segmentation of the target in the scene. The flow is derived as the s
ntationojsea
- Ultrasound images are difficult candidates for shape recovery because of low signal/noise ratio which greatly reduces the observable details within such images. Recently several methods based on evolution of partial differential equations
Tools-for-Tomographic-Reconstruction
- ct重建的matlab程序，可以运行，适合学习和研究使用！希望对大家有用-Ct reconstruction of the matlab program, you can run, suitable for learning and research use! Hope for everyone useful