Francesco Crenna

Panel or jury testing methods in a metrological perspective

Panel or jury testing is a widely used technique in the measurement of quantities related to human perception. However, despite both scientific and applied interests, the confidence of potential users of this technique is less than in more familiar physical and engineering measurements. This lack of confidence arises mainly due to the lack of standardization and of internationally recognized metrological references for panel testing.Here we discuss some methodological aspects of panel testing, including the construction of the reference scale, the definition and realization of a measuring system based on a previously established scale, its calibration and related uncertainty evaluation. We provide illustrative examples concerning the measurement of the quality of sound (acceptability) of machinery used for the automatic sorting of post, presenting experimental results from tests performed in our laboratory, which seem encouraging.Finally, we propose some possible actions for standardization and for the promotion of a metrological approach.

Measurement of quantities depending upon perception by jury-test methods

The measurement of quantities according to perception is of great interest for the optimal characterization of the interaction of persons with their living and working environment. The direct approach to such measurements is based on jury-test methods, which at present have only been partially studied from a metrological perspective and are far from being standardised. So this paper presents a thorough investigation of the jury-test methods, both for the construction of reference measurement scales and for the subsequent development of suitable measurement procedures. The cross-validation of complementary tests is addressed, to ensure the reliability of the results, as well as the importance of test protocols and of operator interfaces. As a support to the proposed methodology, two test cases concerning automotive noises are presented in detail and show good results. This study also intends to promote the standardisation of this kind of method.

Probability as a logic for measurement representations

Measurement can be defined on the basis of the notion of empirical relation, which appears as a primitive term in the representational theory. Here a language is introduced for dealing with empirical relations and a probabilistic logic is applied to statements concerning them.

Hierarchical intelligent measurement set-up for characterizing dynamical systems

Abstract Information technology and distribution instrumentation, supporting the monitoring of dynamical processes, deserve careful consideration to exploit their ability of providing “measurements” with acknowledged metrological consistency. The paper discusses the conceptual background for characterizing vibrating systems through standardized signatures. A hierarchical procedure is devised, based on the recognition of topological, structural and parametrical phases, whose features are singled out and explained by referring to an example instrumental set-up and a case study.

Kinematic measurements for the characterization of the dynamics of hopping

The study of human movement requires the accurate measurement of the forces and of the kinematic quantities involved. Here we consider hopping, that deserves great attention in the literature, since it combines the activation of a complex sensorimotor control with the relative simplicity of the gesture. The movement of the center of mass and the contact conditions are of special interest. Thus a measuring system that combines digital image acquisition, force measurement based on a force plate and extensive use of inertial sensors, for linear and rotational motion, has been developed and characterized. The results concerning the accurate measurement of contact conditions are specially outlined.

Probabilistic inversion: a preliminary discussion

We continue the discussion on the possibility of interpreting probability as a logic, that we have started in the previous IMEKO TC1-TC7-TC13 Symposium. We show here how a probabilistic logic can be extended up to including direct and inverse functions. We also discuss the relationship between this framework and the Bayes-Laplace rule, showing how the latter can be formally interpreted as a probabilistic inversion device. We suggest that these findings open a new perspective in the evaluation of measurement uncertainty.

Instantaneous centre of rotation in human motion: measurement and computational issues

The instantaneous centre of rotation plays an important role in biomechanical modelling and physical-medical interpretation of human gestures. Therefore, we consider its measurement, based on video-image acquisition and processing of human motion records. Measurement and computational aspects are discussed, including the evaluation of measurement uncertainty and the estimation of the effect of some influence quantities on the determination of the position of the instantaneous centre of rotation.