Device to evaluate cleanliness of fiber optic connectors using image processing and neural networks

Abstract

This work proposes a portable, handheld electronic device, which measures the cleanliness in fiber optic connectors via digital image processing and artificial neural networks. Its purpose is to reduce the evaluation subjectivity in visual inspection done by human experts. Although devices with this purpose already exist, they tend to be cost-prohibitive and do not take advantage of neither image processing nor artificial intelligence to improve their results. The device consists of an optical microscope for fiber optic connector analysis, a digital camera adapter, a reduced-board computer, an image processing algorithm, a neural network algorithm and an LCD screen for equipment operation and results visualization. The image processing algorithm applies grayscale histogram equalization, Gaussian filtering, Canny filtering, Hough transform, region of interest segmentation and obtaining radiometric descriptors as inputs to the neural network. Validation consisted of comparing the results by the proposed device with those obtained by agreeing human experts via visual inspection. Results yield an average Cohen's Kappa of 0.926, which implies a very satisfactory performance by the proposed device.

Efficiency of recurrent neural networks for seasonal trended time series modelling

“…ISSN: 2088-8708  Moreover, neural networks (NN) offer new perspectives [12]- [14] for modelling time series than traditional seasonal autoregressive integrated moving average (SARIMA) models [15], [16]. The learning mechanism allows to establish a neural architecture based on parameters such as the size of the input vector, and the number of hidden layers.…”

Automatic optical inspection for detecting keycaps misplacement using Tesseract optical character recognition

“…The preprocessing stage consists of several image-processing steps to obtain binary images. Initially, a color picture is transformed from RGB to grayscale using (3). In this, the variable Y specifies a grayscale image obtained from a colored image with red (R), green (G), and blue (B) channels.…”

“…Many research studies encompass AOI applications on various manufacturing products. For instance, Fernandez et al [3] investigated defects in fiber optic connectors, whereas Li et al [4], Li et al [5], and Rehman et al [6] studied defects in printed circuit boards (PCBs). Moreover, several AOI-based research studies also included consumer products, such as bottles and cans by Rahman et al [7] and Saad et al [8], and keyboards by Huang and Ren [9] and Miao et al [10], [11].…”

Neural network modeling of agglomeration firing process for polymetallic ores

“…A review of the other economic and production areas shows that there is a wide range of neural networks applications and artificial intelligence used for the control, automation, and optimization of the processes and plants in the different areas of control. For example, in the work [26] an encode-decode Seg-Net system via deep learning network incorporated with VGG16 has been proposed in order to identify Int J Elec & Comp Eng ISSN: 2088-8708  the lane markings on distinct environmental effects. Also, there is a recent work on the development of COVID-19 detection system by a transfer deep learning approach where the state-of-the-art CNN models have been used.…”