Today, computer vision inspection systems are widely used for quality control to reduce the costs and to improve the product quality. The glass industry is constantly trying to improve quality by substituting the human control with automated inspection systems but several problems must be solved. In this paper the problem of detecting and measuring the defects of satin glasses is investigated and a real-time system is proposed that is able to analyze the glass surface under inspection, to assess its quality and to characterize its defects. A prototype has been carefully designed and optimized for validating the proposed approach and to reproduce the real issues of quality control. The prototype is composed by several CMOS cameras, a controllable conveyor bend, and an image processing system. Currently the prototype, which is cheap and reliable, is under further development in cooperation with a specialized electronic industry.

The need to correlate as accurately as possible Time to the power quality events pushes towards the necessity to develop circuits able to yield a time reference that is referable to International Time Standards. Moreover, the necessity to deploy distributed power quality monitoring systems requires these time references to be easy to disseminate, making them available on specific monitoring sites. In this paper, the possibility to obtain a time reference continuously traceable to a Cs133 atomic standard is proposed, by locking the Cesium-derived L1 GPS satellites carrier by means of a specifically designed GPS receiver. Such time reference would simply require visibility of one GPS satellite in order to be operative. In particular, the overall architecture of such receiver will be first outlined and then attention will be focused on the design and realization of the VCO which is one of its fundamental building blocks. Measurements carried out on a prototype of the proposed VCO show that its performances are adequate for the aim.

In the paper the author refers to a common opinion about high immunity of current loop measurement and control line to interferences. That opinion is only theoretical. The author presents of some experiments results which are pointed out on limitations concern immunity to interferences especially to the two wire measurement line in 4-20 mA standard.

Magnetic Induction Tomography (MIT) is an imaging technique for passive electrical properties used in industrial and biological imaging. In the case of biological purposes, high resolution measures of the magnetic field induced in the body and the ability to acquire data at several points along its border are two major challenges of MIT. In a previous work we have developed a prototype that allows precisely moving sensing and source coils, improving data quality and quantity for the tomographical image processing. Several alternatives for stability and resolution improvements have been presented. However, all of them had limitations in terms of stability and resolution for a moving coils system. This paper presents a new geometric setup built for a moving sensing coils prototype that allows the cancelation of the carrier field independently of the acquiring position.

The paper considers applying the quadratic joint time – frequency transform for the spectral analysis of the non-stationary magnetic fields. The general properties, advantages and disadvantages of the quadratic algorithms from Cohen’s class like the short – time Fourier transform (STFT) spectrogram and the pseudo Wigner – Ville (PWV) distribution as well as the adaptive spectrogram are discussed. The selected results of the off-line time-frequency analysis of the recorded signals of magnetic field of ship bow thrusters drive with frequency converter are presented. Performed time-frequency analysis enabled the signal identification in time periods. If the frequency contents of the analyzed signal don’t change rapidly, then the STFT spectrogram can be applied with relatively wide window function, to obtain a good frequency resolution. The PWV distribution is useful for analyzing signals that have widely separated components for which a fine time-frequency resolution is required. The adaptive algorithm is characterized by the most accurate detection of investigated nonstationary components, but it requires more computation time, which is suitable only for off-line analysis.