Convolutional neural networks are able to perform a hierarchical learning process starting with local features. However, a limited attention is paid to enhancing such elementary level features like edges. We propose and evaluate two wavelet-based edge-feature enhancement methods to preprocess the input images to convolutional neural networks. The first method develops representations by decomposing the input images using wavelet transform and limited reconstructing subsequently. The second method develops such feature-enhanced inputs to the network using local modulus maxima of wavelet coefficients. For each method, we have developed a new preprocessing layer by implementing each proposed method and have appended to the network architecture. Our empirical evaluations demonstrate that the proposed methods are outperforming the baselines and previously published work.
Microcantilever sensors are commonly used as chemical and biological sensors. Interactions between the functionalisation layer on the cantilever and the analytes in the sample cause the cantilevers to bend. When the analyte concentration is low, these interactions are localized. Then, the same concentration can cause different deflections, depending upon the locations of interactions. The deflections will thus depend on the location of interaction, as well as the concentration of the analyte. This paper presents a model to calculate the deflection, when uni-axial surface stresses are distributed and localized. Results of this model are compared with finite element method simulation results. Biaxial stresses are then considered, and the one dimensional model is shown to be a valid approximation when the stresses are not applied at the ends. Using the model, characteristic response curves of a cantilever, when the surface stresses are localized, are obtained. The probability of determining a concentration based on an observed deflection is shown to be as low as 20%.
Microcantilevers are commonly used as part of sensor elements in Microelectromechanical Systems (MEMS). Deflection or the shift of resonance frequency of microcantilever beams are regularly used to measure chemical, physical or biological quantities. An important characteristic of any sensor is its sensitivity to a given input. This paper explores the possibility of improving the sensitivity of a microcantilever by modifying the mechanical properties using partial perforations on the surface of the microcantilever. This paper presents two analytical models that quantify the deflection and the fundamental resonant frequency in terms of the perforation dimensions for a microcantilever beam. Beams with a single partial perforation are considered first, and the models are then expanded to include multi-perforated cantilevers. Results obtained from the analytical models are compared to Finite Element Analysis (FEA) simulations of perforated microcantilever beams. The analytical models of a microcantilever with a single perforation show high accuracies compared to the FEA, while the accuracy of results for a cantilever beam with many perforations decrease as the number and size of perforations are increased. The results of the models are used to design a cantilever beam with the desired mechanical properties.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.