A Multiple Sensor Detection Design via Embedded Array Optimization and Modified Chebyshev Polynomial Weighting Scheme

In order to meet the demand for odor detection in herbal medicine, a multisensor detection system consisting of various types is implemented. With the consideration of possible imbalance data distribution existing in the herbal medicine datasets, a classifier based on Chebyshev polynomial weighting (CPW) is considered. Besides, the discriminant density is taken into consideration to overcome the negative impact on the calibration recognition result due to the unbalanced number of samples of different classes in the dataset. Moreover, considering the redundancy contained in the sensor data, a sensor array optimization algorithm is explored based on an embedded Hilbert scheme with an area under the curve (AUC) measure. The odor data of five classes of herbal medicine are collected by our system. The effectiveness of the multiple sensor detection system and the proposed algorithm framework show the potential of the system to be applied in the intelligent identification of herbal medicine.