Search results

A purely theoretical approach has been found to be of limited value in the solution of practical Pattern Recognition problems. Difficulties arise when relating infinite mathematics to reality, e.g. “algorithmic convergence” must be replaced by a vaguer notion of “satisfactory performance”. Experimentation has been used to study this and related problems: a) Learning in noise; b) Similarity of classifiers; c) Instability of classifiers; d) Relating infinite‐sample analysis to finite data sets (reference to pdf estimation). Finally, the system requirements for effective experimentation are discussed.

The automatic visual inspection of glassware is a challenging and potentially rewarding field of study. It is possible to detect a wide range of glassware defects and profiles automatically. This article reviews these possibilities and justifies the view that automated visual inspection systems will be increasingly used in glassware inspection tasks.

The paper demonstrates that graphical methods offer some valuable features in the study of probability density function estimators. The design of experiments to augment our existing knowledge of such estimators is outlined. These experiments use artificial data, which is generated with known characteristics. The use of graphical methods in monitoring PDF performance, during work on real problems, is discussed. There is a brief discussion on the use of graphical displays in an interactive system and the paper concludes with a description of a readily comprehended display of classification results.

The problem of pattern classification based upon incomplete vectors was introduced in an earlier paper, where we discussed four possible methods of solution. Of these, only two appeared to be promising as a solution to the general problem. In one of them (a), we envisaged that a multi‐dimensional Monte‐Carlo integration would be needed, while the other procedure (b) required the use of a large store. In this paper, we show how the long integration may be avoided in procedure (a). This is feasible if we use one of a certain class of p.d.f. estimators. We also show that these p.d.f. estimators have certain desirable properties in Method (b).

A depth map module, working with structured light, produces real‐time depth map pictures of three‐dimensional objects.

The automatic visual sensing of aerosol sprays has many advantages over alternative techniques. This article suggests various measurements which could form the basis of an inspection system and illustrates their use on sprays produced by a can of furniture polish. The extension of these ideas to other applications is also discussed.

The fact that you are reading this article indicates that you have optical sensors (eyes). Pick up a pencil from your desk. You will almost certainly locate it visually, guiding your hand towards it by employing visual feedback. Is the tip of the pencil broken? Again, you will probably use optical rather than, say tactile or functional‐testing methods. Over the aeons of our evolutionary history, nature has found that remote sensing by vision is a most valuable facility. It is fast, safer than tactile sensing and can operate over considerable distances. It seems eminently sensible to consider whether these lessons that we learn from nature can be applied in the workshop.

This paper aims to present an analytical investigation of energy and exergy performance on a solar flat plate collector (SFPC) with Cu-CuO/water hybrid nanofluid, Cu/water and CuO/water nanofluids as collector running fluids.

Heat transfer characteristics, pressure drop and energy and exergy efficiencies of SFPC working on these nanofluids are investigated and compared. In this study, a comparison is made by varying the mass flow rates and nanoparticle volume concentration. Thermophysical properties of hybrid nanofluids are estimated using distinctive correlations available in the open literature. Then, the influence of these properties on energy and exergy efficiencies of SFPC is discussed in detail.

Energy analysis reveals that by introducing the hybrid nanoparticles in water, the thermal conductivity of the working fluid is enhanced by 17.52 per cent and that of the individual constituents is enhanced by 15.72 and 15.35 per cent for Cu/water and CuO/water nanofluids, respectively. This resulted in 2.16 per cent improvement in useful heat gain for hybrid nanofluid and 1.03 and 0.91 per cent improvement in heat gain for Cu/water and CuO/water nanofluids, respectively. In line with the above, the collector efficiency increased by 2.175 per cent for the hybrid nanofluid and 0.93 and 1.05 per cent enhancement for Cu/water and CuO/water nanofluids, respectively. Exergy analysis elucidates that by using the hybrid nanofluid, exergy efficiency is increased by 2.59 per cent, whereas it is 2.32 and 2.18 per cent enhancement for Cu/water and CuO/water nanofluids, respectively. Entropy generation is reduced by 3.31, 2.35 and 2.96 per cent for Cu-CuO/water, Cu/water and CuO/water nanofluids, respectively, as compared to water.

However, this is associated with a penalty of increment in pressure drop of 2.92, 3.09 and 2.74 per cent for Cu-CuO/water, Cu/water and CuO/water nanofluids, respectively, compared with water.

It is clear from the analysis that Cu-CuO/water hybrid nanofluids possess notable increment in both energy and exergy efficiencies to use them in SFPCs.

The industrial environment offers many potential applications for image processing, especially in inspection for quality, automatic assembly and the guidance of robotic devices. But development of an appropriate algorithm may require the testing of very many image processing methods. However, a computer system by British Robotic Systems allows easy and interactive testing.

A leading consultant offers advice on the biggest question facing would‐be vision system users: how to choose a system in the first place.