Log-polar transform-based wavelet-modified maximum average correlation height filter for distortion-invariant target recognition

In this paper, we report a log-polar transform-based filter for in-plane rotation and scale-invariant target recognition. The log-polar transform is a known space-invariant image representation used in several image vision systems to eliminate the effects of scale and rotation in an image. In case of in-plane rotation invariance, peaks shift horizontally, while in case of scale invariance, peaks shift vertically. For full out-of-plane rotation-invariance (0–360°), log-polar transformed images are used to train the wavelet-modified maximum average correlation height (WaveMACH) filter. Correlation peak height and peak-to-sidelobe ratio have been calculated as metrics of goodness of the log-polar transform-based WaveMACH filter. This filter would reduce the memory requirement for filter storage in a practical system. Simulation results have been presented.     

Wavelet-modified fringe-adjusted joint transform correlator

In this paper, we implement a wavelet-modified fringe-adjusted joint transform correlator (JTC) for real-time target recognition applications. In real-time situation the input scene is captured using a charge-coupled device (CCD) camera. The obtained joint power spectrum is multiplied by a pre-synthesized fringe-adjusted filter and the resultant function is processed with an appropriately scaled wavelet filter. Three performance measure parameters: correlation peak intensity, peak-to-sidelobe ratio, and signal-to-clutter ratio have been calculated for fringe-adjusted joint transform correlator (FJTC) and wavelet-modified fringe-adjusted joint transform correlator (WFJTC). The WFJTC has been found to yield better results in comparison to conventional FJTC. To suppress the undesired strong dc, the resultant function is differentiated. Differential processing wavelet-modified fringe-adjusted joint power spectrum removes the zero-order spectra and hence improves the detection efficiency. To focus the correlation terms in different planes in order to capture one of the desired autocorrelation peaks and discard the strong dc and another autocorrelation peak, chirp-encoding technique has also been applied. Targets with Gaussian and speckle noise have also been used to check the correlation outputs. Computer simulation and experimental results are presented.     

A multiscale noise tuning stochastic resonance for fault diagnosis in rolling element bearings

Condition monitoring of rotating machinery is important to extend the mechanical system's reliability and operational life. However, in many cases, useful information is often overwhelmed by strong background noise and the defect frequency is difficult to be extracted. Stochastic resonance (SR) is used as a noise-assisted tool to amplify weak signals in nonlinear systems, which can detect weak signals of interest submerged in the noise. The multiscale noise tuning SR (MSTSR), which is originally based on discrete wavelet transform (DWT), has been applied to identify the fault characteristics and has also increased the signal-to-noise ratio (SNR) improvement of SR. Therefore, a novel tri-stable SR method with multiscale noise tuning (MST) is proposed to extract fault signatures for fault diagnosis of rotating machinery. The wavelet packets transform (WPT) based MST can obtain better denoising effect and higher SNR of resonance output compared with the traditional SR method. Thus the proposed method is well-suited for enhancement of rotating machine fault identification, whose effectiveness has been verified by means of practical vibration signals carrying fault information from bearings. Finally, it can be concluded that the proposed method has practical value in engineering.