An Analysis of Ground-Point Classifiers for Terrestrial LiDAR

Previous literature has compared the performance of existing ground point classification (GPC) techniques on airborne LiDAR (ALS) data (LiDAR—light detection and ranging); however, their performance when applied to terrestrial LiDAR (TLS) data has not yet been addressed. This research tested the classification accuracy of five openly-available GPC algorithms on seven TLS datasets: Zhang et al.’s inverted cloth simulation (CSF), Kraus and Pfeiffer’s hierarchical weighted robust interpolation classifier (HWRI), Axelsson’s progressive TIN densification filter (TIN), Evans and Hudak’s multiscale curvature classification (MCC), and Vosselman’s modified slope-based filter (MSBF). Classification performance was analyzed using the kappa index of agreement (KIA) and rasterized spatial distribution of classification accuracy datasets generated through comparisons with manually classified reference datasets. The results identified a decrease in classification accuracy for the CSF and HWRI classification of low vegetation, for the HWRI and MCC classifications of variably sloped terrain, for the HWRI and TIN classifications of low outlier points, and for the TIN and MSBF classifications of off-terrain (OT) points without any ground points beneath. Additionally, the results show that while no single algorithm was suitable for use on all datasets containing varying terrain characteristics and OT object types, in general, a mathematical-morphology/slope-based method outperformed other methods, reporting a kappa score of 0.902.

Download Full-text

Sentiment polarity classification of tweets using a extended dictionary

INTELIGENCIA ARTIFICIAL ◽

10.4114/intartif.vol21iss62pp1-12 ◽

2018 ◽

Vol 21 (62) ◽

pp. 1

Author(s):

Jorge E. Camargo ◽

Vladimir Vargas-Calderon ◽

Nelson Vargas ◽

Liliana Calderón-Benavides

Keyword(s):

Support Vector Machine ◽

Classification Accuracy ◽

Classification Performance ◽

Semantic Relations ◽

Support Vector ◽

The Real ◽

Polarity Classification ◽

Real Academia ◽

Word Definitions

With the purpose of classifying text based on its sentiment polarity (positive or negative), we proposed an extension of a 68,000 tweets corpus through the inclusion of word definitions from a dictionary of the Real Academia Espa\~{n}ola de la Lengua (RAE). A set of 28,000 combinations of 6 Word2Vec and support vector machine parameters were considered in order to evaluate how positively would affect the inclusion of a RAE's dictionary definitions classification performance. We found that such a corpus extension significantly improve the classification accuracy. Therefore, we conclude that the inclusion of a RAE's dictionary increases the semantic relations learned by Word2Vec allowing a better classification accuracy.

Download Full-text

Tree segmentation and classification of deciduous park trees in Sanssouci Park, Potsdam, Germany, using airborne and terrestrial lidar point clouds

10.5194/egusphere-egu21-8689 ◽

2021 ◽

Author(s):

Aljoscha Rheinwalt ◽

Bodo Bookhagen

Keyword(s):

Tree Species ◽

Supervised Classification ◽

Spatial Scales ◽

Point Clouds ◽

Airborne Lidar ◽

Deciduous Tree ◽

Individual Tree ◽

Terrestrial Lidar ◽

Common Tree

While automated, lidar-based tree delineation has proven successful for conifer-dominated forests, deciduous tree stands remain a challenge. &#160;But automatic and reliable segmentation of trees at large spatial scales is a prerequisite for a supervised classification into tree species. We propose an aspect driven tree segmentation that clusters local elevation minima across different aspects. These clusters define tree outlines that respect tree inherent local elevation minima. We validate this approach with more than 25.000 mapped trees of the Sanssouci Park, Potsdam, using an airborne lidar point cloud collected in 2018, and various terrestrial lidar scans for a large fraction of the same park. Further, we demonstrate the tree segmentation by supervised tree species classifications for the most common tree species using random forests and Gaussian process classifiers with geometric parameters derived from individual tree crowns.

Download Full-text

ROI-based features for classification of skin diseases using a multi-layer neural network

Indonesian Journal of Electrical Engineering and Computer Science ◽

10.11591/ijeecs.v23.i1.pp216-228 ◽

2021 ◽

Vol 23 (1) ◽

pp. 216

Author(s):

Thanh-Hai Nguyen ◽

Ba-Viet Ngo

Keyword(s):

Skin Disease ◽

Classification Accuracy ◽

Skin Diseases ◽

Human Life ◽

Disease Diagnosis ◽

Classification Performance ◽

High Classification Accuracy ◽

Separate Region ◽

Occurrence Matrix

Skin diseases have a serious impact on human life and health. This article aims to represent the classification accuracy of skin diseases for supporting the physicians’ correct decision on patients for early treatment. In particular, 100 images in each type of five skin diseases from ISIC database are used for balanced datasets related to the classification accuracy. In addition, this paper focuses on processing images for extracting six optimal types of eleven features of skin disease image for higher classification performance and also this takes less time for training. Therefore, skin disease images are filtered and segmented for separating region of interests (ROIs) before extracting optimal features. First, the skin disease images are processed by normalizing sizes, removing noises, segmenting to separate region of interests (ROIs) showing skin disease signs. Next, a gray-level co-occurrence matrix (GLCM) method is applied for texture analysis to extract eleven features. With the optimal six features chosen, the high classification accuracy of skin diseases is about 92% evaluated using a matrix confusion. The result showed to illustrate the effectiveness of the proposed method. Furthermore, this method can be developed for other medical datasets for supporting in disease diagnosis.

Download Full-text

An Automatic System for Atrial Fibrillation by Using a CNN-LSTM Model

Discrete Dynamics in Nature and Society ◽

10.1155/2020/3198783 ◽

2020 ◽

Vol 2020 ◽

pp. 1-9

Author(s):

Fengying Ma ◽

Jingyao Zhang ◽

Wei Chen ◽

Wei Liang ◽

Wenjia Yang

Keyword(s):

Atrial Fibrillation ◽

Classification Accuracy ◽

Short Term Memory ◽

Sequence Data ◽

Detection System ◽

Heart Rhythm ◽

Classification Performance ◽

Ecg Signals ◽

Electrocardiogram Ecg

Atrial fibrillation (AF) is a common abnormal heart rhythm disease. Therefore, the development of an AF detection system is of great significance to detect critical illnesses. In this paper, we proposed an automatic recognition method named CNN-LSTM to automatically detect the AF heartbeats based on deep learning. The model combines convolutional neural networks (CNN) to extract local correlation features and uses long short-term memory networks (LSTM) to capture the front-to-back dependencies of electrocardiogram (ECG) sequence data. The CNN-LSTM is feeded by processed data to automatically detect AF signals. Our study uses the MIT-BIH Atrial Fibrillation Database to verify the validity of the model. We achieved a high classification accuracy for the heartbeat data of the test set, with an overall classification accuracy rate of 97.21%, sensitivity of 97.34%, and specificity of 97.08%. The experimental results show that our model can robustly detect the onset of AF through ECG signals and achieve stable classification performance, thereby providing a suitable candidate for the automatic classification of AF.

Download Full-text

Viewpoint Analysis for Maturity Classification of Sweet Peppers

Sensors ◽

10.3390/s20133783 ◽

2020 ◽

Vol 20 (13) ◽

pp. 3783

Author(s):

Ben Harel ◽

Rick van Essen ◽

Yisrael Parmet ◽

Yael Edan

Keyword(s):

Random Forest ◽

Classification Accuracy ◽

Sweet Pepper ◽

Classification Performance ◽

Robotic Arm ◽

Random Forest Algorithm ◽

Maturity Level ◽

Depth Images ◽

Manual Classification

The effect of camera viewpoint and fruit orientation on the performance of a sweet pepper maturity level classification algorithm was evaluated. Image datasets of sweet peppers harvested from a commercial greenhouse were collected using two different methods, resulting in 789 RGB—Red Green Blue (images acquired in a photocell) and 417 RGB-D—Red Green Blue-Depth (images acquired by a robotic arm in the laboratory), which are published as part of this paper. Maturity level classification was performed using a random forest algorithm. Classifications of maturity level from different camera viewpoints, using a combination of viewpoints, and different fruit orientations on the plant were evaluated and compared to manual classification. Results revealed that: (1) the bottom viewpoint is the best single viewpoint for maturity level classification accuracy; (2) information from two viewpoints increases the classification by 25 and 15 percent compared to a single viewpoint for red and yellow peppers, respectively, and (3) classification performance is highly dependent on the fruit’s orientation on the plant.

Download Full-text

Classification enhancement of breast cancer histopathological image using penalized logistic regression

Indonesian Journal of Electrical Engineering and Computer Science ◽

10.11591/ijeecs.v13.i1.pp405-410 ◽

2019 ◽

Vol 13 (1) ◽

pp. 405

Author(s):

Mohammed Abdulrazaq Kahya

Keyword(s):

Breast Cancer ◽

Logistic Regression ◽

Classification Accuracy ◽

Area Under The Curve ◽

Classification Performance ◽

Histopathological Image ◽

Classification Of Diseases ◽

Histopathological Images ◽

Penalized Logistic Regression

Classification of breast cancer histopathological images plays a significant role in computer-aided diagnosis system. Features matrix was extracted in order to classify those images and they may contain outlier values adversely that affect the classification performance. Smoothing of features matrix has been proved to be an effective way to improve the classification result via eliminating of outlier values. In this paper, an adaptive penalized logistic regression is proposed, with the aim of smoothing features and provides high classification accuracy of histopathological images, by combining the penalized logistic regression with the smoothed features matrix. Experimental results based on a publicly recent breast cancer histopathological image datasets show that the proposed method significantly outperforms penalized logistic regression in terms of classification accuracy and area under the curve. Thus, the proposed method can be useful for histopathological images classification and other classification of diseases types using DNA gene expression data in the real clinical practice.

Download Full-text

A Study on the Auxiliary Diagnosis of Thyroid Disease Images Based on Multiple Dimensional Deep Learning Algorithms

Current Medical Imaging Formerly Current Medical Imaging Reviews ◽

10.2174/1573405615666190115155223 ◽

2020 ◽

Vol 16 (3) ◽

pp. 199-205

Author(s):

Yuejun Liu ◽

Yifei Xu ◽

Xiangzheng Meng ◽

Xuguang Wang ◽

Tianxu Bai

Keyword(s):

Deep Learning ◽

Learning Algorithms ◽

Region Of Interest ◽

Classification Performance ◽

Thyroid Diseases ◽

Great Success ◽

Learning Models ◽

Good Classification Performance ◽

Spect Images

Background: Medical imaging plays an important role in the diagnosis of thyroid diseases. In the field of machine learning, multiple dimensional deep learning algorithms are widely used in image classification and recognition, and have achieved great success. Objective: The method based on multiple dimensional deep learning is employed for the auxiliary diagnosis of thyroid diseases based on SPECT images. The performances of different deep learning models are evaluated and compared. Methods: Thyroid SPECT images are collected with three types, they are hyperthyroidism, normal and hypothyroidism. In the pre-processing, the region of interest of thyroid is segmented and the amount of data sample is expanded. Four CNN models, including CNN, Inception, VGG16 and RNN, are used to evaluate deep learning methods. Results: Deep learning based methods have good classification performance, the accuracy is 92.9%-96.2%, AUC is 97.8%-99.6%. VGG16 model has the best performance, the accuracy is 96.2% and AUC is 99.6%. Especially, the VGG16 model with a changing learning rate works best. Conclusion: The standard CNN, Inception, VGG16, and RNN four deep learning models are efficient for the classification of thyroid diseases with SPECT images. The accuracy of the assisted diagnostic method based on deep learning is higher than that of other methods reported in the literature.

Download Full-text

Deep Learning-Based Classification of Large-Scale Airborne LiDAR Point Cloud

Canadian Journal of Remote Sensing ◽

10.1080/07038992.2021.1927687 ◽

2021 ◽

pp. 1-15

Author(s):

Mathieu Turgeon-Pelchat ◽

Samuel Foucher ◽

Yacine Bouroubi

Keyword(s):

Deep Learning ◽

Point Cloud ◽

Large Scale ◽

Airborne Lidar

Download Full-text

Artificial Intelligence in Fractured Dental Implant Detection and Classification: Evaluation Using Dataset from Two Dental Hospitals

Diagnostics ◽

10.3390/diagnostics11020233 ◽

2021 ◽

Vol 11 (2) ◽

pp. 233

Author(s):

Dong-Woon Lee ◽

Sung-Yong Kim ◽

Seong-Nyum Jeong ◽

Jae-Hong Lee

Keyword(s):

Dental Implant ◽

Classification Accuracy ◽

Reliability And Validity ◽

Mechanical Complication ◽

Radiographic Images ◽

Acceptable Accuracy ◽

Reliable Detection ◽

Classification Evaluation ◽

Accuracy Performance

Fracture of a dental implant (DI) is a rare mechanical complication that is a critical cause of DI failure and explantation. The purpose of this study was to evaluate the reliability and validity of a three different deep convolutional neural network (DCNN) architectures (VGGNet-19, GoogLeNet Inception-v3, and automated DCNN) for the detection and classification of fractured DI using panoramic and periapical radiographic images. A total of 21,398 DIs were reviewed at two dental hospitals, and 251 intact and 194 fractured DI radiographic images were identified and included as the dataset in this study. All three DCNN architectures achieved a fractured DI detection and classification accuracy of over 0.80 AUC. In particular, automated DCNN architecture using periapical images showed the highest and most reliable detection (AUC = 0.984, 95% CI = 0.900–1.000) and classification (AUC = 0.869, 95% CI = 0.778–0.929) accuracy performance compared to fine-tuned and pre-trained VGGNet-19 and GoogLeNet Inception-v3 architectures. The three DCNN architectures showed acceptable accuracy in the detection and classification of fractured DIs, with the best accuracy performance achieved by the automated DCNN architecture using only periapical images.

Download Full-text

An Imbalanced Image Classification Method for the Cell Cycle Phase

Information ◽

10.3390/info12060249 ◽

2021 ◽

Vol 12 (6) ◽

pp. 249

Author(s):

Xin Jin ◽

Yuanwen Zou ◽

Zhongbing Huang

Keyword(s):

Cell Cycle ◽

Deep Learning ◽

Image Classification ◽

Classification Accuracy ◽

Data Augmentation ◽

Cycle Phase ◽

Generative Adversarial Network ◽

Adversarial Network ◽

Cellular Life

The cell cycle is an important process in cellular life. In recent years, some image processing methods have been developed to determine the cell cycle stages of individual cells. However, in most of these methods, cells have to be segmented, and their features need to be extracted. During feature extraction, some important information may be lost, resulting in lower classification accuracy. Thus, we used a deep learning method to retain all cell features. In order to solve the problems surrounding insufficient numbers of original images and the imbalanced distribution of original images, we used the Wasserstein generative adversarial network-gradient penalty (WGAN-GP) for data augmentation. At the same time, a residual network (ResNet) was used for image classification. ResNet is one of the most used deep learning classification networks. The classification accuracy of cell cycle images was achieved more effectively with our method, reaching 83.88%. Compared with an accuracy of 79.40% in previous experiments, our accuracy increased by 4.48%. Another dataset was used to verify the effect of our model and, compared with the accuracy from previous results, our accuracy increased by 12.52%. The results showed that our new cell cycle image classification system based on WGAN-GP and ResNet is useful for the classification of imbalanced images. Moreover, our method could potentially solve the low classification accuracy in biomedical images caused by insufficient numbers of original images and the imbalanced distribution of original images.

Download Full-text