Detecting Shipping Container Impacts with Vertical Cell Guides inside Container Ships during Handling Operations.

Due to the mechanical nature of container handling operations, as well as natural factors, container and handling infrastructure suffers various types of damage during use, especially within the tight and enclosed environments of a ship's hull. In this operational environment, it is critical to detect any sort of physical impacts between the vertical cell guides of the ship's hull and the container. Currently, an inspection of impacts and evaluation of any consequences is performed manually, via visual inspection processes. This process is time-consuming and relies on the technical expertise of the personnel involved. In this paper, we propose a five-step impact-detection methodology (IDM), intended to detect only the most significant impact events based on acceleration data. We conducted real measurements in a container terminal using a sensory device placed on the spreader of the quay crane. The proposed solution identified an average of 12.8 container impacts with the vertical cell guides during common handling operations. In addition, the results indicate that the presented IDM can be used to recognize repeated impacts in the same space of each bay of the ship, and can be used as a decision support tool for predictive maintenance systems.

Pattern Recognition Methods and Features Selection for Speech Emotion Recognition System.

The impact of the classification method and features selection for the speech emotion recognition accuracy is discussed in this paper. Selecting the correct parameters in combination with the classifier is an important part of reducing the complexity of system computing. This step is necessary especially for systems that will be deployed in real-time applications. The reason for the development and improvement of speech emotion recognition systems is wide usability in nowadays automatic voice controlled systems. Berlin database of emotional recordings was used in this experiment. Classification accuracy of artificial neural networks, k-nearest neighbours, and Gaussian mixture model is measured considering the selection of prosodic, spectral, and voice quality features. The purpose was to find an optimal combination of methods and group of features for stress detection in human speech. The research contribution lies in the design of the speech emotion recognition system due to its accuracy and efficiency.