Correct Extraction Rate Research Articles

Mapping 10 m monthly surface water dynamics in the Yangtze River basin from 2017 to 2020 using a robust ATMC algorithm

Surface water bodies are irreplaceable in the global climate and ecological system. However, quantifying the extent and change of these water bodies has been challenging due to environmental complexity, climatic diversity, and their highly dynamic nature in seasons and interannual periods. Most established remote-sensing water extraction methods involve a practical trade-off between spatial and temporal resolution. To remedy these deficiencies, we proposed a simple and robust Automatic Threshold–based Multi-source data Combination algorithm to improve the efficiency and accuracy of water mapping, achieving monthly stable periodic monitoring. The algorithm fully considered the principles and advantages of Sentinel-1/2 images for water detection and performed adaptive optimal threshold calculation based on the Edge Otsu algorithm. A new product of Surface Water Dynamics in the Yangtze River Basin was generated on a monthly temporal scale and 10-m spatial scale from 2017 to 2020 using the algorithm based on the Google Earth Engine platform. The confusion matrix-based metrics based on sample points were 95.85 % (producer's accuracy), 99.16 % (user's accuracy), 97.41 % (overall accuracy), 0.95 (Matthews correlation coefficient), and 0.95 (Kappa coefficient). Furthermore, the overall correct extraction rate of water bodies can reach 95.85 % and 91.87 % for small water bodies, with commission and omission errors of 0.04 % and 0.03 % at the sub-pixel scale, respectively. The distribution of water bodies within the basin is highly uneven, and their changes exhibit a distinct seasonal trend. The maximum inundation extent occurred in August 2020 (summer, 53,372.88 km2), while the minimum inundation extent was observed in November 2019 (winter, 42,491.74 km2). The successful application of the algorithm in the Yangtze River Basin established a good benchmark for surface water monitoring of multi-source satellite observations and showed great potential for global-scale applications. It will assist the monitoring and evaluation services of the UN Sustainable Development Goals and provide solutions to more global and regional issues.

LBP feature and hash function based dual watermarking algorithm for database

In this paper, we propose a local binary pattern (LBP) feature and hash function based dual watermarking algorithm for database. Attribute feature columns are selected to generate zero watermarks using the Pearson correlation method. The zero watermarks are generated by the LBP. The attribute values of the selected feature columns are divided into two parts for embedding and extracting the watermark. Zero watermark feature code is embedded into the lowest significant bit of the selected attribute column by using the database watermarking algorithm with hash function for copyright authentication. The proposed method uses two-layer hashing method to improve the robustness. In the watermark extraction, a method of combining majority voting and Hamming error correction code is used to control the watermark error to ensure the correct extraction rate of the watermark. Experimental results show that the algorithm not only provides good availability of the database after embedding the watermark, but also ensures the correct extraction of the watermark after various attacks.

Remote-Sensing Extraction of Small Water Bodies on the Loess Plateau

The mixed pixel of low-resolution remote-sensing image makes the traditional water extraction method not effective for small water body extraction. This study takes the Loess Plateau with complex terrain as the research area and develops a multi-index fusion threshold segmentation algorithm (MFTSA) for a large-scale small water body extraction algorithm based on GEE (Google Earth Engine). MFTSA uses the AWEI (automated water extraction index), MNDWI (modified normalized difference water index), NDVI (normalized difference vegetation index) and EVI (enhanced vegetation index) for multi-index synergy to extract small water bodies. It also uses slope data generated by the SRTM (Shuttle Radar Topography Mission digital elevation model) and NIR band reflectance to eliminate suppressing high reflectivity noise and shadow noise. An MFTSA algorithm was proposed and the results showed that: (1) The overall extraction accuracy of the MFTSA algorithm on the Loess Plateau was 98.14%, and the correct extraction rate of small water bodies was 92.82%. (2) Compared with traditional water index methods and classification methods, the MFTSA algorithm could extract small water bodies with higher integrity and clearer and more accurate boundaries. (3) The MFTSA algorithm was used to extract a total of 69,900 small water bodies on the Loess Plateau, accounting for 97.63% of the total water bodies, and the area was 482.11 square kilometers, accounting for 16.50% of the total water bodies.

A risk factor tracing method for LNG receiving terminals based on GAT and a bidirectional LSTM network

LNG terminals are important transfer stations for LNG storage, processing, and outbound transmission which involve intensive manual operation and safety management activities. Traditional safety evaluation methods can provide "cause-effect" information when investigating the traceability of incorrect manual operation. However, manual analysis and summarizing risk factor causality is time-consuming and labor-intensive, and there are inaccuracy problems. Usually, HAZOP meetings require safety, electrical, mechanical, instrumentation and other engineers to brainstorm and summarize the risk factors of LNG receiving terminals. This process is laborious and time-consuming, and there is no guarantee that the person will be able to record it completely every time. In this paper, a graph attention network (GAT) and bidirectional Long and Short-Term Memory Network (LSTM) deep integrated learning model is proposed. The model can automatically extract risk factor causal relationships. After starting with a risk factor text, the number type features and syntactic dependency graph features of the input text are fused to enhance the causal relationship traceability of incorrect personnel operation. Taking the unloading and storage processes as examples, in terms of the effect of automatic extraction risk factors, the integrated algorithm model improves the extraction accuracy by up to 7.1% compared with a single algorithm. In addition, the correct extraction rate of different cause-effect relationships increases by 21.57%, while the integrated algorithm considers sentence matching, improves the granularity of cause-effect information based on the traditional method, expands the root cause of risk factors, and the traceability results are consistent with on-site accident event results. At the same time, it is more conducive to accurately managing incorrect manual operations. In the model training sample size sensitivity analysis, the algorithm has a superior correct causal label extraction rate for different sample sizes when the number of hidden layer stacking layers is K= 2, which solves the problem of model adaptation for the text volume of LNG terminal risk factors.

Crop Row Detection in the Middle and Late Periods of Maize under Sheltering Based on Solid State LiDAR

As the basic link of autonomous navigation in agriculture, crop row detection is vital to achieve accurate detection of crop rows for autonomous navigation. Machine vision algorithms are easily affected by factors such as changes in field lighting and weather conditions, and the majority of machine vision algorithms detect early periods of crops, but it is challenging to detect crop rows under high sheltering pressure in the middle and late periods. In this paper, a crop row detection algorithm based on LiDAR is proposed that is aimed at the middle and late crop periods, which has a good effect compared with the conventional machine vision algorithm. The algorithm proposed the following three steps: point cloud preprocessing, feature point extraction, and crop row centerline detection. Firstly, dividing the horizontal strips equally, the improved K-means algorithm and the prior information of the previous horizontal strip are utilized to obtain the candidate points of the current horizontal strip, then the candidate points information is used to filter and extract the feature points in accordance with the corresponding threshold, and finally, the least squares method is used to fit the crop row centerlines. The experimental results show that the algorithm can detect the centerlines of crop rows in the middle and late periods of maize under the high sheltering environment. In the middle period, the average correct extraction rate of maize row centerlines was 95.1%, and the average processing time was 0.181 s; in the late period, the average correct extraction rate of maize row centerlines was 87.3%, and the average processing time was 0.195 s. At the same time, it also demonstrates accuracy and superiority of the algorithm over the machine vision algorithm, which can provide a solid foundation for autonomous navigation in agriculture.

Robust Automatic Segmentation of Inflamed Appendix from Ultrasonography with Double-Layered Outlier Rejection Fuzzy C-Means Clustering

Accurate diagnosis of acute appendicitis from abdominal ultrasound is a challenging task, since traditional sonographic diagnostic criteria for appendicitis, such as diameter, compressibility, and wall thickness, rely on complete identification or visualization of the appendix and the diagnosis is frequently operator subjective. In this paper, we propose a robust automatic segmentation method for inflamed appendix identification to mitigate abovementioned difficulties. We use outlier rejection fuzzy c-means clustering (FCM) algorithm within a double-layered learning structure to extract the target inflamed appendix area. The proposed method extracts the target appendix in 98 cases out of 100 test images, which is far better than traditional FCM, standard outlier FCM, and double-layered learning with FCM in correct extraction rate. Furthermore, we investigate the outlier rejection effect and double layered learning effect by comparing our proposed method with standard double-layered FCM and the standard outlier-rejection FCM. In this comparison, the proposed method exhibits robust segmentation results in accuracy, precision, and recall by 2.5~5.6% over two standard methods in quality with human pathologists’ marking as the ground truth.

AUTOMATIC EXTRACTION OF POWER LINES FROM UAV LIDAR POINT CLOUDS USING A NOVEL SPATIAL FEATURE

Abstract. UAV LiDAR systems have unique advantage in acquiring 3D geo-information of the targets and the expenses are very reasonable; therefore, they are capable of security inspection of high-voltage power lines. There are already several methods for power line extraction from LiDAR point cloud data. However, the existing methods either introduce classification errors during point cloud filtering, or occasionally unable to detect multiple power lines in vertical arrangement. This paper proposes and implements an automatic power line extraction method based on 3D spatial features. Different from the existing power line extraction methods, the proposed method processes the LiDAR point cloud data vertically, therefore, the possible location of the power line in point cloud data can be predicted without filtering. Next, segmentation is conducted on candidates of power line using 3D region growing method. Then, linear point sets are extracted by linear discriminant method in this paper. Finally, power lines are extracted from the candidate linear point sets based on extension and direction features. The effectiveness and feasibility of the proposed method were verified by real data of UAV LiDAR point cloud data in Sichuan, China. The average correct extraction rate of power line points is 98.18%.

Long-distance sea wave extraction method based on improved Otsu algorithm

Sea wave extraction is the foundation of sea wave measurement, which plays a significant role in binocular stereo vision-based tsunami early-warning systems. However, long-distance sea surface images possess two typical features: ununiformed illuminance and a low signal-to-noise ratio that makes it difficult to extract sea waves out from sea surface images and that leads to incorrect extractions. In this paper, we improved the Otsu algorithm to realize a dynamic multi-threshold algorithm for sea wave extraction. First, the original image was divided into small blocks and the dynamic multi-thresholds were determined according to local sea surface illuminance. Then an evaluation function was defined to add constraint conditions to adjacent blocks’ thresholds. Finally, sea waves were extracted out using the blocks’ thresholds in all the blocks. As seen in our experimental results, the correct extraction rate was 80.9%, most sea waves are extracted regardless of their intensity.

Word Extraction from Arabic Handwritten Documents Based on Statistical Measures

In Arabic, word extraction is particularly challenging because words are often divided into sub-words, and a few letters do not connect to the following letter. In this paper, we present an efficient method for extracting words from Arabic handwritten documents. The proposed method is based on two groups of spatial measures (the lengths of connected components (CCs) and the gaps between these CCs) which differentiate successive CCs in text lines. Lengths are clustered into three distinct clusters to identify an optimal threshold for separating isolated letters, sub-words, and words. Besides, Gaps are clustered into two clusters, to indicate whether the gap occurs between-words or within-a word. This clustering is implemented using Self-Organizing Map (SOM) algorithm. The efficiency of the proposed method was tested by conducting experiments on 35 ages of handwritten Arabic text, accessed from benchmarking Database for Arabic Handwritten Text Recognition Research (AHDB). Our tests produced very promising results, achieving a correct extraction rate of 86.3%.

Secret Communication Based on Quantization Tables

Because of the prevalence of internet, the information security issues have attracted much attention in recent years. Protecting privacy of exchanging information through media becomes the subject of many studies. The Quantization in JPEG is to encode the coefficients transformed by Discrete Cosine Transform (DCT) using a Quantization Table (QT) generated by a quality factor. Quantizing DCT coefficients by different QTs can be adopted to record the secret messages. Based on the variation of QTs, we propose two schemes in frequency domain to improve a data hiding method, in terms of correct extraction rate and image quality. Experimental results show that the proposed approaches can provide better correct extraction rate, higher image quality, and satisfactory embedding capacity.

Road Edges Detection of a Novel Travel Aid for Visual Impaired People

A road edges detection method of a novel electronic travel aid for the blind was proposed. Road edges extraction was realized by Canny filter and Hough line transfer. Multiple lines detected by Hough transfer were processed into one edge line. Road deviation judgment method was proposed to determine if the user has deviated from the road by judging the position and angle of the edge lines. Experiments showed that the correct extraction rate of road edges was 93%, and the road deviation judgment method was effective and fast. Experiments were done using the consecutive stereo image sequences captured on road.

A character segmentation algorithm for mixed‐mode communication

AbstractIn the realization of a mixed‐mode communication, it is necessary before recognizing individual characters, to separate text and black and white figure regions and to extract efficiently the character region. Problems in such a procedure are the detection and correction of the inclination of a document, separation of contact characters, the merge of disconnected characters, and extraction of a letterhead. This paper describes the results of studies on such problems. The document considered is the English text image, containing binary figures and a letterhead. The basic idea is as follows. Connected regions are obtained by directional propagation and shrinking to merge figures. Then: (1) a thinning process is performed to detect the inclination angle of the input text; (2) the sizes of the connected regions and their relative locations are examined to extract the letterhead. Estimation of pitch is performed and statistical data about individual characters are used to separate contact characters or merge disconnected characters. The experiment was made for 10 and 12 pitch printed characters, and the correct extraction rate was 100 percent for individual characters, and 94.6 percent for a letterhead. Thus it was verified that the proposed method is useful in extracting the character region.