Use MATLAB® and Simulink® to gain insight into your image and video data, develop algorithms, and explore implementation tradeoffs.
- Design vision solutions with a comprehensive set of reference-standard algorithms for image processing, computer vision, and deep learning.
- Collaborate with teams using OpenCV, Python, and C/C++ using interoperable APIs and integration tools.
- Use workflow apps to automate common tasks and accelerate algorithm exploration.
- Accelerate algorithms on NVIDIA GPUs, cloud, and datacenter resources without specialized programming or IT knowledge.
- Deploy algorithms to embedded devices, including NVIDIA GPUs, Intel processors and FPGAs, and ARM-based embedded processors.
Using MATLAB for Image Processing and Computer Vision
Image Apps and Visualization
Use MATLAB apps to explore your data interactively and automatically generate MATLAB code. This means you don’t have to code from scratch. Explore the following featured apps:
- Camera Calibration
Estimate camera intrinsics, extrinsics, and lens distortion parameters.
- Image and Video Labeling
Label ground truth in a collection of images, and view videos and image sequences.
- Image Segmentation
Segment an image using active contours and graph cutting algorithms such as grabcut and lazy snapping.
Apps for Visualization
Identify and extract meaningful information from images and videos.
- Volume Visualization
View 3D volumetric data as volumes or as plane slices with the Volume Viewer App
- Video Viewer
Select the movie or image sequence that you want to play, jump to a specific frame in the sequence, or change the frame rate of the display.
- DICOM Browser
Explore a collection of DICOM files, select and import into MATLAB.
- Volume Visualization App (2:19)
Image Processing and Computer Vision Applications
Perform a wide range of image processing and computer vision tasks directly from MATLAB. These include:
- 3D image processing workflows
- What Is Object Detection? (3:20)
- Image segmentation and registration
- Point cloud processing
- Stereo vision
Integration with Open Source
Integrate directly with open source. You can reuse legacy code written in another programming language, create MATLAB powered responsive web sites, or program hardware using error-free embedded C-code generated directly from MATLAB.
Direct Camera Access and Image and Video Import
Connect to cameras through hardware support packages. You can acquire live images and video from frame grabbers, GigE Vision® cameras, DCAM cameras, and more.
MATLAB supports standard data and image formats, and you can access your data with prebuilt functions and apps. Import and manage large datasets not able to fit into memory with
Parallelize workflows using multi-core CPUs or NVIDIA GPUS without reprogramming algorithms.
Run MATLAB on the cloud or in your browser. And with Parallel Computing Toolbox™, you can solve computationally and data-intensive problems using multicore processors, GPUs, and computer clusters.
With MATLAB, you can work with C/C++ and HDL code. Run image processing algorithms on PC hardware, FPGAs, and ASICs, and develop imaging systems.
GPU Coder™ generates optimized CUDA® code from MATLAB code for deep learning, embedded vision, and autonomous systems. You can use the generated CUDA within MATLAB to accelerate computationally intensive portions of your MATLAB code.