Machine Vision and Perception

image: Laser welding system

Machine vision control of a laser welding process.

DepthX - Autonomous underwater robot

SwRI engineers developed embedded vision systems for robot navigation and sample analysis to detect signs of life for a project funded through NASA.

Depainting with a laser.

SwRI engineers developed a high-speed vision system for providing real-time monitoring and control of a laser coating removal process.

Machine or computer vision can provide comparable, and sometimes superior functionality to that of the human eye - distinguishing among thousands of hues, providing sensitivity beyond the visible spectrum, perceiving objects in three dimensions (3D) - and more. Using visible and infrared cameras, 3D scanners, structured lighting, and other technologies, computer vision systems automatically acquire and analyze data to solve real world problems. Data analysis techniques range from classical image processing and metrology operations to advanced machine learning methods.

Machine Vision Applications

Machine vision enables robots to autonomously navigate or to perform precision movements. Image analysis and pattern recognition algorithms locate tumors in the body, identify customers at bank automatic teller machines, and detect hidden cracks in machined parts. The range of computer vision applications is practically endless.

Machine Vision Technologies

SwRI engineers have extensive experience developing state of the art software and hardware for machine vision applications . SwRI will perform system design tasks and can integrate commercial off-the-shelf (COTS) components or develop custom systems. With expertise in advanced computer technology, Southwest Research Institute (SwRI) conducts customer-driven research and applied engineering projects in a wide range of machine vision technologies, including:

  • Image processing
  • Pattern recognition
  • Scene analysis
  • Classification and grading
  • 3D imaging
  • Multispectral imaging
  • Robot guidance
  • Process control
  • Industrial inspection
  • Change detection
  • Model-based vision
  • Sensor fusion
  • Dimensional inspection
  • Motion analysis
  • Infrared image analysis
  • Object recognition

SwRI develops image analysis systems that operate in harsh industrial environments, monitor process quality, prevent process failures, and measure product quality. These systems provide guidance for precision assembly, robotics, dimensional inspection, and defect detection.

Industrial Automation and Quality Control

Machine vision systems can provide quality control and real-time feedback for industrial processes, overcoming physical limitations and subjective judgment of humans. Automated systems capable of high-speed measurement of parts or a process enable manufacturers to achieve unprecedented product quality.

Automated Precision Measurement

Computer vision techniques automate measuring tasks that require high precision and accuracy. The resulting measurements are used to position and orient parts, provide numerical data for subsequent analysis (modeling and simulation), or monitor the performance and efficiency of mechanical devices. An automated dimensional measurement system is a fast, accurate way to measure the size, shape, or profile of many components.

Related Terminology

computer vision  •  machine vision  •  image processing  •  scene analysis  •  pattern recognition  •  model-based vision  •  sensor fusion  •  motion analysis  •  infrared image analysis  •  dimensional inspection  •  industrial inspection  •   process control  •  perception  •  robot guidance  •  3D imaging  •  multispectral imaging  •  object recognition

Benefiting government, industry and the public through innovative science and technology
Southwest Research Institute® (SwRI®), headquartered in San Antonio, Texas, is a multidisciplinary, independent, nonprofit, applied engineering and physical sciences research and development organization with 9 technical divisions.