Search

Based on convolution neural network, this paper proposed a new asymmetric two branch underwater biological classification model to solve the problems of low visibility, poor illumination conditions and no obvious differences among species in the underwater environment. In the model, the interactive branch used different convolution neural network to extracted local features and interacted with local features through the interactive module to enhance the classification model. Convolutional neural network branch could effectively learn the global characteristics of the target and made up for the global information ignored in the interactive branch. Finally, this model obtained 98 . 9%, 98 . 3% and 97 . 9% of the accuracy on the three data sets of Fish4-Knowledge (F4K), Eilat and RAMAS, which were significantly improved compared with the previous methods. Visual interpretation also verifies that the model could effectively capture local features and eliminated the background influence. Finally, it shows that the model has good classification performance in underwater environment. [ABSTRACT FROM AUTHOR]

As the frequency of operation of recent mobile communication devices has increased and their bandwidth has expanded, the behavior of electromagnetic waves inside the devices has attracted attention. In the microstrip lines used for these devices, T-junctions are used for branching in general. This structure is more likely to be affected by the transmission line orientation. In contrast, the asymmetric branch is less restrictive. On the other hand, the transmitted power is divided in an unbalanced way in high-frequency regions. In this paper, the unbalance of the transmitted power at the asymmetrical branch is quantitatively evaluated and a solution is discussed. It is found that a notch provided near the branch is effective. FDTD analysis reveals that the transmitted power unbalance of 16.2% at 10 GHz is reduced to 2.3% by means of a notch at the asymmetrical branch. Measurement confirms the effectiveness of our method. © 2005 Wiley Periodicals, Inc. Electron Comm Jpn Pt 1, 88(5): 20–27, 2005; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/ecja.20164 [ABSTRACT FROM AUTHOR]