The development of process automation and the consequent increase of productivity are currently associated with the better accuracy of quality control systems thus representing one of the most important prerogatives to acquire the minimum levels of competitiveness in the market. Machine Vision must be considered as one of the most efficient and used solutions, thanks to its applicability and integration with different production sectors.
The Loccioni Group in collaboration with robot producers has realized a series of solutions for quality testing in the production line using video cameras and anthropomorphic robots, able to check a high number of tests in a short cycle time.
Stereo vision is a 3D reconstruction method based on the triangulation principle.
Since it uses 2 cameras, structured light is not needed, thus it is faster since the depth information is evaluated using only the left and right image.
The flexibility of the stereo system can be increased by applying the variable baseline technique in which the distance between the two cameras (the baseline) is changed by using a displacement device such as a motorized stage or a robotized arm.
A profilometer is composed by a camera and a laser projecting a pattern, generally a line. When the line impacts an object with different thickness it will be deflected in a proportional way to such variations of height. The leaps the pattern laser undergoes are evaluated starting from the two-dimension image. The section is extracted and analysed to extract the third dimension; in this way measurements of depth, width, area can be carried out.
Optical Coherence Tomography
OCT is an optical imaging technique that is fast, very sensitive and non-destructive. It provides a resolution on the micron scale and a cross-sectional images with an high penetration depth.
Hyperspectral imaging collects and processes information from across the electromagnetic spectrum. The goal of hyperspectral imaging is to obtain the spectrum for each pixel in the image of a scene, with the purpose of finding objects or identifying materials. In fact, certain objects leave unique ‘fingerprints’ in the electromagnetic spectrum. Known as spectral signatures, these ‘fingerprints’ enable identification of the materials that make up a scanned object. Hyperspectral sensors and processing systems have been used for applications in food and pharmaceutical sectors.
Catadioptric Vision means acquiring images with a conventional camera through an optical system composed by a mirror and a standard lens.
Catadioptric vision allows imaging cilindrical surfaces in just one shot, without displacing the imaging system or rotating the object.
Colorimetry is the science and technology used to quantify and describe physically the human color perception.
Usually colorimetry is performed as a single point measurement with dedicated instruments.
Sometimes colorimetric inspection must be done on a bigger area and such a specifically developed calibrated vision system is used.
The stereophotometer use the Shape From Shading technique (reconstruction of three‑dimension sections starting from shade analysis).
The illuminator is composed of a dome-light to diffuse the light and of N leds switching on in sequence and projecting shadows on the surfaces; every time the leds switch on an image is acquired by the camera located in the middle.
The camera, therefore, always focuses the same zone but with different lighting conditions. By knowing the light inclination angle and the brightness variation it is possible to trace the defect dimension.
Infrared thermography (IRT), thermal imaging, and thermal video are examples of infrared imaging science. Thermographic cameras detect radiation in the infrared range of the electromagnetic spectrum and produce images of that radiation, called thermograms. The amount of radiation emitted by an object increases with temperature; therefore, thermography allows one to see variations in temperature.
High Speed Machine Vision with FPGA Image Processing
Latest applications more and more require powerful embedded processing architecture for real-time high-performance machine vision. FPGA-based custom computing machines have been shown to provide the performance required for real-time image acquisition and processing and now is widely adopted in different fields of application.