Wyniki wyszukiwania - BazTech

1

Automated lung nodule detection in CT images by optimized CNN: impact of improved whale optimization algorithm

Kumar M. Kiran, Amalanathan Anthoniraj

Computer Assisted Methods in Engineering and Science

|

2022

|

Vol. 29, no. 1-2 spec.

7--31

EN

Lung cancer is one of the leading causes of cancer-related deaths among individuals.It should be diagnosed at the early stages, otherwise it may lead to fatality due to itsmalicious nature. Early detection of the disease is very significant for patients’ survival, andit is a challenging issue. Therefore, a new model including the following stages: (1) imagepre-processing, (2) segmentation, (3) proposed feature extraction and (4) classificationis proposed. Initially, pre-processing takes place, where the input image undergoes specificpre-processing. The pre-processed images are then subjected to segmentation, which iscarried out using the Otsu thresholding model. The third phase is feature extraction, wherethe major contribution is obtained. Specifically, 4D global local binary pattern (LBP)features are extracted. After their extracting, the features are subjected to classification,where the optimized convolutional neural network (CNN) model is exploited. For a moreprecise detection of a lung nodule, the filter size of a convolution layer, hidden unit inthe fully connected layer and the activation function in CNN are tuned optimally byan improved whale optimization algorithm (WOA) called the whale with tri-level enhancedencircling behavior (WTEEB) model.

2

The Impact and Importance of Fabric Image Preprocessing for the New Method of Individual Inter-Thread Pores Detection

Owczarek Magdalena

Autex Research Journal

|

2020

|

Vol. 20, no. 3

250--262

EN

The paper describes the impact and importance of preprocessing methods of fabric image for detection of inter-thread pores (ITP), which is a new method of individual ITP identification. The aim of this experiment is to identify precisely every individual ITP of fabric structure by using optimal preprocessing algorithm for further quantitative, morphometric structural analysis of specialized fabrics (barriers, industrial filters, composites, others) in context of air permeability, flow resistance, UV radiation, viruses penetration, thermal comfort by estimation fabric porosity, especially macroporosity parameters and cover factor. The correct identification of individual ITP depends on the acquisition method and the preprocessing algorithm. It was conducted by analyzing the adaptation of digital image preprocessing methods for two structures of plain weave fabric in two magnification zooms, 1.25 and 0.8. Preprocessing operations were performed in the area of spatial operations of the image. The optimal preprocessing algorithm includes low-pass filtering, histogram equalization, nonlinear filtering, thresholding, and morphological operation. This algorithm was selected based on the factors developed by the author (ITP detection, RID factor—a difference between the real and model ITP areas) which rely on the ITP size, shape, and location. The graphic view of the ITP contour position on the fabric image is a verification element in the optimal preprocessing algorithm. The presented results of the air permeability of two different plain weave structures confirm the need to optimize the algorithm of pre-image processing methods to precisely detect each individual ITP in the fabric image.

3

Preprocessing Techniques for Online Signature Verification and Identification

Pień A., AdamskI M.

Advances in Computer Science Research

|

2018

|

Nr 14

163--176

EN

Handwritten signature is a behavioral biometric that can be used for automatic signer verification and identification. Online signature, in addition to visual shape, incorporates dynamics of the writing process such as trajectory, velocity and additional characteristics such as pen pressure and angles. While there are many approaches to online signature verification proposed in the literature, only few works related to preprocessing and its effect on the system performance. In this work selected preprocessing techniques were investigated such as: normalization, noise filtering and resampling. The evaluation was carried out in verification and identification tasks based on DTW distance measure and signatures from SVC2004 database.

PL

Podpis odręczny jest behawioralną cechą biometryczną która umożliwia automatyczną weryfikację i identyfikację autora podpisu. Podpis dynamiczny, oprócz informacji o kształcie, zawiera również dane dotyczące dynamiki składania podpisu takie jak trajektoria kreślenia, prędkość, zmiana nacisku i kątów nachylenia pióra. W literaturze można znaleźć wiele podejść do automatycznej weryfikacji podpisu, brakuje jednak prac z szerszą analizą metod wstępnego przetwarzania i oceną ich wpływu na poprawność pracy całego systemu. W niniejszej pracy zbadano wybrane techniki wstępnego przetwarzania takie jak: normalizacja, filtracja, próbkowanie oraz oceniono ich użyteczność w procesie weryfikacji i identyfikacji podpisu. W badaniach wykorzystano system bazujący na mierze odległości Dynamic Time Warping. Eksperymenty przeprowadzono na podpisach dynamicznych z bazy SVC2004.