La traducción de esta página aún no se ha actualizado a la versión más reciente. Haga clic aquí para ver la última versión en inglés.

Procesamiento de imágenes mediante deep learning

Realice tareas de procesamiento de imágenes, como eliminar el ruido de la imagen y realice una traslación entre imágenes, mediante técnicas de deep learning que usan redes neuronales profundas (requiere Deep Learning Toolbox™)

Las técnicas de deep learning usan redes neuronales para aprender representaciones útiles de ciertas características directamente a partir de los datos. Por ejemplo, se puede usar una red neuronal previamente entrenada para identificar y eliminar ciertos artefactos de una imagen, como el ruido.

A noisy image is passed to a trained denoising network, which returns a denoised image.

Funciones

expandir todo

Crear almacenes de datos para el preprocesamiento de imágenes

`augmentedImageDatastore`	Transformar lotes para aumentar datos de imágenes
`blockedImageDatastore`	Datastore for use with blocks from `blockedImage` objects (desde R2021a)
`denoisingImageDatastore`	Denoising image datastore
`imageDatastore`	Datastore for image data
`randomPatchExtractionDatastore`	Datastore for extracting random 2-D or 3-D random patches from images or pixel label images
`transform`	Transform datastore
`combine`	Combine data from multiple datastores

Aumentar imágenes

`jitterColorHSV`	Randomly alter color of pixels
`randomWindow2d`	Randomly select rectangular region in image (desde R2021a)
`randomCropWindow3d`	Create randomized cuboidal cropping window
`centerCropWindow2d`	Crear una ventana de recorte rectangular central
`centerCropWindow3d`	Crear una ventana de recorte cuboidal central
`Rectangle`	Extensiones espaciales de una región rectangular 2D
`Cuboid`	Extensiones espaciales de una región cuboidal 3D
`randomAffine2d`	Crear una transformación afín 2D aleatoria
`randomAffine3d`	Create randomized 3-D affine transformation
`affineOutputView`	Crear una vista de salida para deformar imágenes
`imerase`	Remove image pixels within rectangular region of interest (desde R2021a)

Cambiar el tamaño y la forma de las entradas de deep learning

`resize2dLayer`	2-D resize layer (desde R2020b)
`resize3dLayer`	3-D resize layer (desde R2020b)
`dlresize`	Resize spatial dimensions of `dlarray` object (desde R2020b)
`DepthToSpace2DLayer`	Depth to space layer (desde R2021a)
`SpaceToDepthLayer`	Space to depth layer (desde R2020b)
`depthToSpace`	Rearrange `dlarray` data from depth dimension into spatial blocks (desde R2021a)
`spaceToDepth`	Rearrange spatial blocks of `dlarray` data along depth dimension (desde R2021a)

Crear redes de deep learning

`encoderDecoderNetwork`	Create encoder-decoder network (desde R2021a)
`blockedNetwork`	Create network with repeating block structure (desde R2021a)
`pretrainedEncoderNetwork`	Create encoder network from pretrained network (desde R2021a)
`cycleGANGenerator`	Create CycleGAN generator network for image-to-image translation (desde R2021a)
`patchGANDiscriminator`	Create PatchGAN discriminator network (desde R2021a)
`pix2pixHDGlobalGenerator`	Create pix2pixHD global generator network (desde R2021a)
`addPix2PixHDLocalEnhancer`	Add local enhancer network to pix2pixHD generator network (desde R2021a)
`unitGenerator`	Create unsupervised image-to-image translation (UNIT) generator network (desde R2021a)
`unitPredict`	Perform inference using unsupervised image-to-image translation (UNIT) network (desde R2021a)

Eliminar el ruido de una imagen

`denoiseImage`	Eliminar el ruido de una imagen mediante una red neuronal profunda
`denoisingNetwork`	Obtener una red para eliminar el ruido de una imagen
`dnCNNLayers`	Get denoising convolutional neural network layers

Temas

Preprocesar datos de imágenes para deep learning

Get Started with Image Preprocessing and Augmentation for Deep Learning

Preprocess data for deep learning applications with deterministic operations such as resizing, or augment training data with randomized operations such as random cropping.
Datastores for Deep Learning (Deep Learning Toolbox)
Learn how to use datastores in deep learning applications.
Prepare Datastore for Image-to-Image Regression (Deep Learning Toolbox)
This example shows how to prepare a datastore for training an image-to-image regression network using the transform and combine functions of ImageDatastore.

Crear redes neuronales para aplicaciones de procesamiento de imágenes

Train and Apply Denoising Neural Networks
Use a pretrained neural network to remove Gaussian noise from a grayscale image, or train your own network using predefined layers.
Create Modular Neural Networks
You can create and customize deep learning networks that follow a modular pattern with repeating groups of layers, such as U-Net and cycleGAN.
Get Started with GANs for Image-to-Image Translation
Transfer styles and characteristics from one set of images to the scene content of other images by using generative adversarial networks (GANs).
Redes neuronales profundas preentrenadas (Deep Learning Toolbox)
Aprenda a descargar y utilizar redes neuronales convolucionales preentrenadas para clasificación, transferencia del aprendizaje y extracción de características.
Lista de capas de deep learning (Deep Learning Toolbox)
Descubra todas las capas de deep learning de MATLAB^®.

Deep learning en MATLAB

Deep learning en MATLAB (Deep Learning Toolbox)
Descubra las prestaciones de deep learning en MATLAB utilizando redes neuronales convolucionales para clasificación y regresión, incluidas redes preentrenadas y transferencia del aprendizaje, así como entrenamiento en unidades de procesamiento gráfico (GPU), unidades CPU, clusters y nubes.
Semantic Segmentation Using Deep Learning (Computer Vision Toolbox)
This example shows how to segment an image using a semantic segmentation network.

Ejemplos destacados

Increase Image Resolution Using Deep Learning

Create a high-resolution image from a low-resolution image using a very-deep super-resolution (VDSR) neural network.

Abrir script en vivo

Unsupervised Day-to-Dusk Image Translation Using UNIT

Translate images between daytime and dusk lighting conditions using an unsupervised image-to-image translation network (UNIT).

Abrir script en vivo

Detect Image Anomalies Using Explainable FCDD Network

Use an anomaly detector to distinguish between normal pills and pills with anomalous chips or contamination.

Abrir script en vivo

Classify Defects on Wafer Maps Using Deep Learning

Classify eight types of manufacturing defects on wafer maps using a simple convolutional neural network (CNN).

Abrir script en vivo

Detect Image Anomalies Using Pretrained ResNet-18 Feature Embeddings

Train a similarity-based anomaly detector using one-class learning of feature embeddings extracted from a pretrained ResNet-18 convolutional neural network.

Abrir script en vivo

Unsupervised Medical Image Denoising Using CycleGAN

Generate high-quality high-dose computed tomography (CT) images from noisy low-dose CT images using a CycleGAN neural network.

Abrir script en vivo

Unsupervised Medical Image Denoising Using UNIT

Generate high-quality computed tomography (CT) images from noisy low-dose CT images using a UNIT neural network.

Abrir script en vivo

JPEG Image Deblocking Using Deep Learning

Reduce JPEG compression artifacts in an image using a denoising convolutional neural network (DnCNN).

Abrir script en vivo

Image Processing Operator Approximation Using Deep Learning

Approximate an image filtering operation using a multiscale context aggregation network (CAN).

Abrir script en vivo

Develop Camera Processing Pipeline Using Deep Learning

Convert RAW camera data to an aesthetically pleasing color image using a U-Net.

Abrir script en vivo

Brighten Extremely Dark Images Using Deep Learning

Recover brightened RGB images from RAW camera data collected in extreme low-light conditions using a U-Net.

Abrir script en vivo

Classify Tumors in Multiresolution Blocked Images

Classify multiresolution whole slide images (WSIs) that might not fit in memory using an Inception-v3 deep neural network.

Abrir script en vivo

3-D Brain Tumor Segmentation Using Deep Learning

Perform semantic segmentation of brain tumors from 3-D medical images.

Abrir script en vivo

Semantic Segmentation of Multispectral Images Using Deep Learning

Perform semantic segmentation of a multispectral image with seven channels using U-Net.

Abrir script en vivo

Neural Style Transfer Using Deep Learning

Apply the stylistic appearance of one image to the scene content of a second image using a pretrained VGG-19 network.

Abrir script en vivo

Remove Noise from Color Image Using Pretrained Neural Network

Remove Gaussian noise from an RGB image using a denoising convolutional neural network.

Abrir script en vivo

Quantify Image Quality Using Neural Image Assessment

Analyze the aesthetic quality of images using a Neural Image Assessment (NIMA) convolutional neural network (CNN).

Abrir script en vivo