Signature image processing - Wikipedia

Signature Image Processing (SIP) is a technology used for weld quality assurance. Implementations of the process analyze electrical data collected from welding processes. Acceptable welding requires exact conditions; variations in conditions can cause a weld to be unacceptable. SIP allows welding faults to be identified in real time, measures the stability of welding processes, and enables welding processes to be optimized.

Stephen Simpson at the University of Sydney conceived of the idea of using electrical data analyzed by algorithms to assess the quality of the welds produced in robotic manufacturing in 1995. The method stemmed from research on the complex physical phenomena that occur in welding arcs. The critical realization was that a way of determining the quality of a weld could be developed without a definitive understanding of those phenomena.^[1][2][3] The development involved a number of advances. The first was a method for handling sampled data blocks by treating them as phase-space portrait signatures with appropriate image processing. Typically, one second's worth of sampled welding voltage and current data are collected from GMAW pulse or short arc welding processes. The data is converted to a 2D histogram, and signal-processing operations such as image smoothing are performed.^[4] The second advance was a technique for analyzing welding signatures based on statistical methods from the social sciences, such as principal component analysis. The relationship between the welding voltage and the current reflects the state of the welding process, and this information is contained in the signature image. Comparing signatures quantitatively using principal component analysis allows for the spread of signature images, enabling faults to be detected^[5] and identified^[6] The system includes algorithms and mathematics appropriate for real-time welding analysis on personal computers, and the multidimensional optimization of fault-detection performance using experimental welding data.^[7] Comparing signature images from moment to moment in a weld provides a useful estimate of how stable the welding process is.^[8][9] "Through-the-arc" sensing, by comparing signature images when the physical parameters of the process change, leads to quantitative estimates—for example, of the position of the weld bead.^[10]

Unlike systems that log information for later study or that use X-rays or ultrasound to check samples, this technology looks at the electrical signal and detects faults when they occur.^[11] Data blocks of 4,000 points of electrical data are collected four times a second and converted to signature images. After image processing operations, statistical analyses of the signatures provide quantitative assessment of the welding process, revealing its stability and reproducibility, and providing fault detection and process diagnostics.^[12] A similar approach, using voltage-current histograms and a simplified statistical measure of distance between signature images has been evaluated for tungsten inert gas (TIG) welding by researchers from Osaka University.^[13]

SIP is the basis for the WeldPrint system (owned by the University of Sydney). WeldPrint was developed with the assistance of an Australian government R&D Start grant (1999–2001), after support by the Australian Research Council for the fundamental research (1997–2001). The system consists of a front-end interface and software based on the SIP engine, and relies on electrical signals alone. It is designed to be non-intrusive and sufficiently robust to withstand harsh industrial welding environments. The first major purchaser of the technology, GM Holden^[14][15][16] provided feedback that allowed the system to be refined in ways that increased its industrial and commercial value. Improvements in the algorithms, including multiple parameter optimization with a server network, have led to an order-of-magnitude improvement in fault-detection performance over the past five years.^{[citation needed]}

The WeldPrint software received the Brother business software of the year award (2001); in 2003, the technology received the A $100,000 inaugural Australasian Peter Doherty Prize for Innovation;^[17][18] and WTi, the University's original spin-off company, received an AusIndustry Certificate of Achievement in recognition of the development.^{[citation needed]}

SIP has opened opportunities for researchers to use it as a measurement tool both in welding^[19] and in related disciplines, such as structural engineering.^[20] Research opportunities have opened up in the application of biomonitoring of external EEGs, where SIP offers advantages in interpreting the complex signals^[21]