optimization toolbox matlab 2020b  (MathWorks Inc)

Journal: Communications Engineering

Article Title: Towards industry-ready additive manufacturing: AI-enabled closed-loop control for 3D melt electrowriting

doi: 10.1038/s44172-024-00302-4

Figure Lengend Snippet: A User selects printing properties such as the fibre size, deposition accuracy, printing speed and printing stability, which refer to fibre diameter, jet angle, collection speed and standard deviation of the Taylor cone area respectively. B User input is passed through the ANN optimizer—the feedforward model of the system which predicts a set of input parameters that will satisfy the user’s output parameter selection. C Printing commences using the predicted input parameters. The material deposition process is captured in real-time using the collimated backlight, telecentric lens, and monochrome camera. D Output images are collected and analysed. E Output data is displayed and used in ( B ) the feedback loop where the ANN controller is used to correct the discrepancy between the requested and measured output. F System architecture: MATLAB is used for data operations and user interface, while the microcontroller boards and camera are utilized for parameter control and output monitoring.

Article Snippet: Neural network training was performed using MATLAB 2020b and the Deep Learning Toolbox.

Techniques: Standard Deviation, Selection, Control

Journal: Procedia Computer Science

Article Title: Grape Leaf Disease Diagnosis System Using Fused Deep Learning Features Based System

doi: 10.1016/j.procs.2024.04.037

Figure Lengend Snippet: Fig. 2. The methods recommended processing flow Third, combine the deep characteristics that were retrieved using either concatenation of canonical correlation analyses, CCA sum, or straight concatenation. The deep-fused characteristics should then be fed into a fine-trained SVM classifier, which may subsequently output the various diseases present in the input grape leaves. A multi-class SVM classifier may be trained using the “appropriate class output codes for error correction. “MATLAB 2020b's (fitcecoc) function is used using its built-in parameterization. The "fitcecoc" function uses a model of binary SVM with a one-to-one encoding architecture to boost the classifier's categorization effectiveness. Table 2 displays some of the default settings used throughout the training phase. Table 2 Several common training-stage parameters. Component Rate

Article Snippet: A MATLAB 2020b software environment is used, which, by installing the Deep Learning toolbox, may support several traditional CNN models, including AlexNet, GooLeNet, and ResNet.

Techniques: