scholarly journals Automatic Changes Detection between Outdated Building Maps and New VHR Images Based on Pre-Trained Fully Convolutional Feature Maps

Sensors ◽  
2020 ◽  
Vol 20 (19) ◽  
pp. 5538
Author(s):  
Yunsheng Zhang ◽  
Yaochen Zhu ◽  
Haifeng Li ◽  
Siyang Chen ◽  
Jian Peng ◽  
...  
Keyword(s):  
Cross Validation ◽  
High Spatial Resolution ◽  
Data Cleaning ◽  
Poor Performance ◽  
Uncertainty Estimation ◽  
Feature Maps ◽  
Existing Building ◽  
Feature Extractor ◽  
Changes Detection ◽  
Initial Classification

Detecting changes between the existing building basemaps and newly acquired high spatial resolution remotely sensed (HRS) images is a time-consuming task. This is mainly because of the data labeling and poor performance of hand-crafted features. In this paper, for efficient feature extraction, we propose a fully convolutional feature extractor that is reconstructed from the deep convolutional neural network (DCNN) and pre-trained on the Pascal VOC dataset. Our proposed method extract pixel-wise features, and choose salient features based on a random forest (RF) algorithm using the existing basemaps. A data cleaning method through cross-validation and label-uncertainty estimation is also proposed to select potential correct labels and use them for training an RF classifier to extract the building from new HRS images. The pixel-wise initial classification results are refined based on a superpixel-based graph cuts algorithm and compared to the existing building basemaps to obtain the change map. Experiments with two simulated and three real datasets confirm the effectiveness of our proposed method and indicate high accuracy and low false alarm rate.

scholarly journals Evaluation of Sampling and Cross-Validation Tuning Strategies for Regional-Scale Machine Learning Classification

2019 ◽  
Vol 11 (2) ◽  
pp. 185 ◽  
Author(s):  
Christopher A. Ramezan ◽  
Timothy A. Warner ◽  
Aaron E. Maxwell
Keyword(s):  
Machine Learning ◽  
Monte Carlo ◽  
Spatial Resolution ◽  
Cross Validation ◽  
High Spatial Resolution ◽  
Regional Scale ◽  
Sample Selection ◽  
Selection Methods ◽  
Machine Learning Classification ◽  
Leave One Out

High spatial resolution (1–5 m) remotely sensed datasets are increasingly being used to map land covers over large geographic areas using supervised machine learning algorithms. Although many studies have compared machine learning classification methods, sample selection methods for acquiring training and validation data for machine learning, and cross-validation techniques for tuning classifier parameters are rarely investigated, particularly on large, high spatial resolution datasets. This work, therefore, examines four sample selection methods—simple random, proportional stratified random, disproportional stratified random, and deliberative sampling—as well as three cross-validation tuning approaches—k-fold, leave-one-out, and Monte Carlo methods. In addition, the effect on the accuracy of localizing sample selections to a small geographic subset of the entire area, an approach that is sometimes used to reduce costs associated with training data collection, is investigated. These methods are investigated in the context of support vector machines (SVM) classification and geographic object-based image analysis (GEOBIA), using high spatial resolution National Agricultural Imagery Program (NAIP) orthoimagery and LIDAR-derived rasters, covering a 2,609 km2 regional-scale area in northeastern West Virginia, USA. Stratified-statistical-based sampling methods were found to generate the highest classification accuracy. Using a small number of training samples collected from only a subset of the study area provided a similar level of overall accuracy to a sample of equivalent size collected in a dispersed manner across the entire regional-scale dataset. There were minimal differences in accuracy for the different cross-validation tuning methods. The processing time for Monte Carlo and leave-one-out cross-validation were high, especially with large training sets. For this reason, k-fold cross-validation appears to be a good choice. Classifications trained with samples collected deliberately (i.e., not randomly) were less accurate than classifiers trained from statistical-based samples. This may be due to the high positive spatial autocorrelation in the deliberative training set. Thus, if possible, samples for training should be selected randomly; deliberative samples should be avoided.

High spatial resolution and non-destructive evaluation of wood density and microfibril angle by NIR hyperspectral imaging

NIR news ◽  
2017 ◽  
Vol 28 (5) ◽  
pp. 7-12 ◽  
Author(s):  
Te Ma ◽  
Tetsuya Inagaki ◽  
Satoru Tsuchikawa
Keyword(s):  
Hyperspectral Imaging ◽  
Wood Density ◽  
Spatial Resolution ◽  
Cross Validation ◽  
High Spatial Resolution ◽  
Microfibril Angle ◽  
Coefficient Of Determination ◽  
X Ray ◽  
Effective Manner ◽  
Non Destructive

Wood density and microfibril angle are strongly correlated with wood stiffness, shrinkage, and anisotropy. Understanding the spatial distribution of these values is critical for solid timber applications. In this study, near infrared (NIR) hyperspectral imaging was used to evaluate wood density and microfibril angle in a non-destructive, yet effective manner. Briefly, five wood samples collected from both normal and compression parts of two different Cryptomeria japonica trees were analyzed. Partial least squares regression analysis was performed to determine the relationship between X-ray reference data and NIR spectra, and cross-validation (leave-one-out) was used for checking prediction performances. The validation coefficient of determination (r2) between predicted densities by the NIR technique and measured values by SilviScan (X-ray data) was 0.83 with a root mean squared error of cross-validation (RMSECV) of 105.18 kg/m3. Regarding microfibril angle, r2 and RMSECV were 0.77 and 5.36°, respectively. Finally, wood density and microfibril angle were successfully mapped at a high spatial resolution (156 µm) to facilitate the detection of annual growth ring features and evaluation of aspects of heterogeneous wood quality.

scholarly journals Deep Distillation Recursive Network for Remote Sensing Imagery Super-Resolution

2018 ◽  
Vol 10 (11) ◽  
pp. 1700 ◽  
Author(s):  
Kui Jiang ◽  
Zhongyuan Wang ◽  
Peng Yi ◽  
Junjun Jiang ◽  
Jing Xiao ◽  
...  
Keyword(s):  
Remote Sensing ◽  
Video Processing ◽  
Satellite Images ◽  
State Of The Art ◽  
Satellite Image ◽  
Super Resolution ◽  
Poor Performance ◽  
Feature Maps ◽  
Deep Convolutional Neural Networks ◽  
Frequency Components

Deep convolutional neural networks (CNNs) have been widely used and achieved state-of-the-art performance in many image or video processing and analysis tasks. In particular, for image super-resolution (SR) processing, previous CNN-based methods have led to significant improvements, when compared with shallow learning-based methods. However, previous CNN-based algorithms with simple direct or skip connections are of poor performance when applied to remote sensing satellite images SR. In this study, a simple but effective CNN framework, namely deep distillation recursive network (DDRN), is presented for video satellite image SR. DDRN includes a group of ultra-dense residual blocks (UDB), a multi-scale purification unit (MSPU), and a reconstruction module. In particular, through the addition of rich interactive links in and between multiple-path units in each UDB, features extracted from multiple parallel convolution layers can be shared effectively. Compared with classical dense-connection-based models, DDRN possesses the following main properties. (1) DDRN contains more linking nodes with the same convolution layers. (2) A distillation and compensation mechanism, which performs feature distillation and compensation in different stages of the network, is also constructed. In particular, the high-frequency components lost during information propagation can be compensated in MSPU. (3) The final SR image can benefit from the feature maps extracted from UDB and the compensated components obtained from MSPU. Experiments on Kaggle Open Source Dataset and Jilin-1 video satellite images illustrate that DDRN outperforms the conventional CNN-based baselines and some state-of-the-art feature extraction approaches.

scholarly journals Vehicle Logo Recognition Based on Enhanced Matching for Small Objects, Constrained Region and SSFPD Network

Sensors ◽  
2019 ◽  
Vol 19 (20) ◽  
pp. 4528 ◽  
Author(s):  
Ruikang Liu ◽  
Qing Han ◽  
Weidong Min ◽  
Linghua Zhou ◽  
Jianqiang Xu
Keyword(s):  
Recognition Performance ◽  
Recognition Rate ◽  
Poor Performance ◽  
Supplementary Information ◽  
Single Shot ◽  
Feature Maps ◽  
Logo Recognition ◽  
Region Extraction ◽  
Feature Pyramid ◽  
Vehicle Logo Recognition

Vehicle Logo Recognition (VLR) is an important part of vehicle behavior analysis and can provide supplementary information for vehicle identification, which is an essential research topic in robotic systems. However, the inaccurate extraction of vehicle logo candidate regions will affect the accuracy of logo recognition. Additionally, the existing methods have low recognition rate for most small vehicle logos and poor performance under complicated environments. A VLR method based on enhanced matching, constrained region extraction and SSFPD network is proposed in this paper to solve the aforementioned problems. A constrained region extraction method based on segmentation of the car head and car tail is proposed to accurately extract the candidate region of logo. An enhanced matching method is proposed to improve the detection performance of small objects, which augment each of training images by copy-pasting small objects many times in the unconstrained region. A single deep neural network based on a reduced ResNeXt model and Feature Pyramid Networks is proposed in this paper, which is named as Single Shot Feature Pyramid Detector (SSFPD). The SSFPD uses the reduced ResNeXt to improve classification performance of the network and retain more detailed information for small-sized vehicle logo detection. Additionally, it uses the Feature Pyramid Networks module to bring in more semantic context information to build several high-level semantic feature maps, which effectively improves recognition performance. Extensive evaluations have been made on self-collected and public vehicle logo datasets. The proposed method achieved 93.79% accuracy on the Common Vehicle Logos Dataset and 99.52% accuracy on another public dataset, respectively, outperforming the existing methods.

scholarly journals Learning Attention-Aware Interactive Features for Fine-Grained Vegetable and Fruit Classification

2021 ◽  
Vol 11 (14) ◽  
pp. 6533
Author(s):  
Yimin Wang ◽  
Zhifeng Xiao ◽  
Lingguo Meng
Keyword(s):  
Interaction Network ◽  
Superior Performance ◽  
Feature Interaction ◽  
Human Beings ◽  
Feature Maps ◽  
Neural Structure ◽  
Global Features ◽  
Data Set ◽  
Fine Grained ◽  
Feature Extractor

Vegetable and fruit recognition can be considered as a fine-grained visual categorization (FGVC) task, which is challenging due to the large intraclass variances and small interclass variances. A mainstream direction to address the challenge is to exploit fine-grained local/global features to enhance the feature extraction and representation in the learning pipeline. However, unlike the human visual system, most of the existing FGVC methods only extract features from individual images during training. In contrast, human beings can learn discriminative features by comparing two different images. Inspired by this intuition, a recent FGVC method, named Attentive Pairwise Interaction Network (API-Net), takes as input an image pair for pairwise feature interaction and demonstrates superior performance in several open FGVC data sets. However, the accuracy of API-Net on VegFru, a domain-specific FGVC data set, is lower than expected, potentially due to the lack of spatialwise attention. Following this direction, we propose an FGVC framework named Attention-aware Interactive Features Network (AIF-Net) that refines the API-Net by integrating an attentive feature extractor into the backbone network. Specifically, we employ a region proposal network (RPN) to generate a collection of informative regions and apply a biattention module to learn global and local attentive feature maps, which are fused and fed into an interactive feature learning subnetwork. The novel neural structure is verified through extensive experiments and shows consistent performance improvement in comparison with the SOTA on the VegFru data set, demonstrating its superiority in fine-grained vegetable and fruit recognition. We also discover that a concatenation fusion operation applied in the feature extractor, along with three top-scoring regions suggested by an RPN, can effectively boost the performance.

scholarly journals Authentication of Transylvanian Spirits Based on Isotope and Elemental Signatures in Conjunction with Statistical Methods

Foods ◽  
2021 ◽  
Vol 10 (12) ◽  
pp. 3000
Author(s):  
Dana Alina Magdas ◽  
Gabriela Cristea ◽  
Adrian Pîrnau ◽  
Ioana Feher ◽  
Ariana Raluca Hategan ◽  
...  
Keyword(s):  
Statistical Methods ◽  
Mineral Content ◽  
Cross Validation ◽  
Practical Importance ◽  
Linear Discriminant ◽  
Potential Association ◽  
Good Classification ◽  
Validation Procedure ◽  
The Cross ◽  
Initial Classification

The potential association between stable isotope ratios of light elements and mineral content, in conjunction with unsupervised and supervised statistical methods, for differentiation of spirits, with respect to some previously defined criteria, is reviewed in this work. Thus, based on linear discriminant analysis (LDA), it was possible to differentiate the geographical origin of distillates in a percentage of 96.2% for the initial validation, and the cross-validation step of the method returned 84.6% of correctly classified samples. An excellent separation was also obtained for the differentiation of spirits producers, 100% in initial classification, and 95.7% in cross-validation, respectively. For the varietal recognition, the best differentiation was achieved for apricot and pear distillates, a 100% discrimination being obtained in both classifications (initial and cross-validation). Good classification percentages were also obtained for plum and apple distillates, where models with 88.2% and 82.4% in initial and cross-validation, respectively, were achieved for plum differentiation. A similar value in the cross-validation procedure was reached for the apple spirits. The lowest classification percent was obtained for quince distillates (76.5% in initial classification followed by 70.4% in cross-validation). Our results have high practical importance, especially for trademark recognition, taking into account that fruit distillates are high-value commodities; therefore, the temptation of “fraud”, i.e., by passing regular distillates as branded ones, could occur.

scholarly journals Application of Advanced Land Observing Satellite 3 (ALOS-3) Data to Land Cover and Vegetation Mapping

2021 ◽  
Author(s):  
Ram C. Sharma ◽  
Hidetake Hirayama ◽  
Keitarou Hara
Keyword(s):  
Land Cover ◽  
Cross Validation ◽  
High Spatial Resolution ◽  
National Level ◽  
Kappa Coefficient ◽  
Boosted Decision Tree ◽  
Cross Track ◽  
Along Track ◽  
Fold Cross Validation ◽  
Wide Swath

Advanced Land Observing Satellite 3 (ALOS-3) is capable of observing global land areas with wide swath (4000 km along-track direction and 70 km cross-track direction) at high spatial resolution (panchromatic: 0.8m, multispectral: 3.2m). Maintenance and updating of Land Cover and Vegetation (LCV) information at national level is one of the major goals of the ALOS-3 mission. This paper presents the potential of simulated ALOS-3 images for the classification and mapping of LCV types. We simulated WorldView-3 images according to the configuration of the ALOS-3 satellite sensor and the ALOS-3 simulated (ALOS-3S) images were utilized for the classification and mapping of LCV types in two cool temperate ecosystems. This research dealt with classification and mapping of 17 classes in the Hakkoda site and 25 classes in the Zao site. We employed a Gradient Boosted Decision Tree (GBDT) classifier with 10-fold cross-validation method for assessing the potential of ALOS-3S images. In the Hakkoda site, we obtained overall accuracy, 0.811 and kappa coefficient, 0.798. In the Zao site, overall accuracy and kappa coefficient were 0.725 and 0.711 respectively. Regardless of limited temporal scenes available in the research, ALOS-3S images showed high potential (at least 0.711 kappa-coefficient) for the LCV classification. The availability of more temporal scenes from ALOS-3 satellite is expected for improved classification and mapping of LCV types in the future.

scholarly journals Instance Segmentation Based on Deep Convolutional Neural Networks and Transfer Learning for Unconstrained Psoriasis Skin Images

2021 ◽  
Vol 11 (7) ◽  
pp. 3155
Author(s):  
Guo-Shiang Lin ◽  
Kuan-Ting Lai ◽  
Jian-Ming Syu ◽  
Jen-Yung Lin ◽  
Sin-Kuo Chai
Keyword(s):  
Neural Networks ◽  
Transfer Learning ◽  
Convolutional Neural Networks ◽  
Cross Validation ◽  
Normal Skin ◽  
Object Localization ◽  
Feature Maps ◽  
Deep Convolutional Neural Networks ◽  
Subjective Testing ◽  
Instance Segmentation

In this paper, an efficient instance segmentation scheme based on deep convolutional neural networks is proposed to deal with unconstrained psoriasis images for computer-aided diagnosis. To achieve instance segmentation, the You Only Look At CoefficienTs (YOLACT) network composed of backbone, feature pyramid network (FPN), Protonet, and prediction head is used to deal with psoriasis images. The backbone network is used to extract feature maps from an image, and FPN is designed to generate multiscale feature maps for effectively classifying and localizing objects with multiple sizes. The prediction head is used to predict the classification information, bounding box information, and mask coefficients of objects. Some prototypes generated by Protonet are combined with mask coefficients to estimate the pixel-level shapes for objects. To achieve instance segmentation for unconstrained psoriasis images, YOLACT++ with a pretrained model is retrained via transfer learning. To evaluate the performance of the proposed scheme, unconstrained psoriasis images with different severity levels are collected for testing. As for subjective testing, the psoriasis regions and normal skin areas can be located and classified well. The four performance indices of the proposed scheme were higher than 93% after cross validation. About object localization, the Mean Average Precision (mAP) rates of the proposed scheme were at least 85.9% after cross validation. As for efficiency, the frames per second (FPS) rate of the proposed scheme reached up to 15. In addition, the F1_score and the execution speed of the proposed scheme were higher than those of the Mask Region-Based Convolutional Neural Networks (R-CNN)-based method. These results show that the proposed scheme based on YOLACT++ can not only detect psoriasis regions but also distinguish psoriasis pixels from background and normal skin pixels well. Furthermore, the proposed instance segmentation scheme outperforms the Mask R-CNN-based method for unconstrained psoriasis images.

scholarly journals Structural neuroimaging biomarkers for obsessive-compulsive disorder in the ENIGMA-OCD consortium: medication matters

2020 ◽  
Vol 10 (1) ◽  
Author(s):  
Willem B. Bruin ◽  
◽  
Luke Taylor ◽  
Rajat M. Thomas ◽  
Jonathan P. Shock ◽  
...  
Keyword(s):  
Obsessive Compulsive Disorder ◽  
Cross Validation ◽  
Model Performance ◽  
Poor Performance ◽  
Structural Mri ◽  
Classification Performance ◽  
Brain Anatomy ◽  
Obsessive Compulsive ◽  
Compulsive Disorder ◽  
Disease Marker

Abstract No diagnostic biomarkers are available for obsessive-compulsive disorder (OCD). Here, we aimed to identify magnetic resonance imaging (MRI) biomarkers for OCD, using 46 data sets with 2304 OCD patients and 2068 healthy controls from the ENIGMA consortium. We performed machine learning analysis of regional measures of cortical thickness, surface area and subcortical volume and tested classification performance using cross-validation. Classification performance for OCD vs. controls using the complete sample with different classifiers and cross-validation strategies was poor. When models were validated on data from other sites, model performance did not exceed chance-level. In contrast, fair classification performance was achieved when patients were grouped according to their medication status. These results indicate that medication use is associated with substantial differences in brain anatomy that are widely distributed, and indicate that clinical heterogeneity contributes to the poor performance of structural MRI as a disease marker.

scholarly journals THE USE OF NEURAL NETWORKS IN RARE VEGETATION COMMUNITIES CLASSIFICATION

2014 ◽  
pp. 171-184
Author(s):  
E. M. Gambarova
Keyword(s):  
Multilayer Perceptron ◽  
Classification Scheme ◽  
High Spatial Resolution ◽  
Statistical Tests ◽  
Classification Schemes ◽  
Geoinformation Technologies ◽  
Training Samples ◽  
Community Types ◽  
Neural Classifier ◽  
Initial Classification

This paper describes training of Multilayer Perceptron Neural classifier to extract rare vegetation objects from high spatial resolution IKONOS satellite imagery. There have been considered three options of training of the Multilayer Perceptron Neural according to three different classification schemes. At first 12 type of rare vegetation community types were defined, a main classification scheme (“Initial classification scheme”) was designed on that base. After prelim statistical tests on training samples two modification algorithms of the classification scheme were defined: the first one led to creating of scheme consisting of 7 classes (“Modified classification scheme”) and second one led us to creating of 5-classes scheme (“Optimized classification scheme“). The learning procedures of these classifiers are described as well as analysis and post processing of extraction results of objects of interest using Geoinformation Technologies in details.

Sign in / Sign up

Export Citation Format

Share Document