Abstract: This paper describes a methodology for remote
performance monitoring of retail refrigeration systems. The proposed
framework starts with monitoring of the whole refrigeration circuit
which allows detecting deviations from expected behavior caused by
various faults and degradations. The subsequent diagnostics methods
drill down deeper in the equipment hierarchy to more specifically
determine root causes. An important feature of the proposed concept
is that it does not require any additional sensors, and thus, the
performance monitoring solution can be deployed at a low
installation cost. Moreover only a minimum of contextual
information is required, which also substantially reduces time and
cost of the deployment process.
Abstract: In this paper, we were introduces a skin detection
method using a histogram approximation based on the mean shift
algorithm. The proposed method applies the mean shift procedure to a
histogram of a skin map of the input image, generated by comparison
with standard skin colors in the CbCr color space, and divides the
background from the skin region by selecting the maximum value
according to brightness level. The proposed method detects the skin
region using the mean shift procedure to determine a maximum value
that becomes the dividing point, rather than using a manually selected
threshold value, as in existing techniques. Even when skin color is
contaminated by illumination, the procedure can accurately segment
the skin region and the background region. The proposed method may
be useful in detecting facial regions as a pretreatment for face
recognition in various types of illumination.
Abstract: In this paper, a novel method for recognition of musical
instruments in a polyphonic music is presented by using an
embedded hidden Markov model (EHMM). EHMM is a doubly
embedded HMM structure where each state of the external HMM
is an independent HMM. The classification is accomplished for
two different internal HMM structures where GMMs are used as
likelihood estimators for the internal HMMs. The results are compared
to those achieved by an artificial neural network with two
hidden layers. Appropriate classification accuracies were achieved
both for solo instrument performance and instrument combinations
which demonstrates that the new approach outperforms the similar
classification methods by means of the dynamic of the signal.