In this paper, the positional error curve of point features was extended to an error curves band of line segment features. Firstly, the constitution and shape of the error curves band were analyzed. On this basis, the...In this paper, the positional error curve of point features was extended to an error curves band of line segment features. Firstly, the constitution and shape of the error curves band were analyzed. On this basis, the general boundary curve formula of that band was derived. Secondly, the visualizing error curves bands were realized through three exam- ples. Finally,area index has been examined by comparing numerical results from error curves band and error ellipes band.展开更多
By adopting the method of controlling parameters this paper describes the construction of various kinds of cubic curve segment and curved surface fragment with rational and non rational parameters, and discusses the ...By adopting the method of controlling parameters this paper describes the construction of various kinds of cubic curve segment and curved surface fragment with rational and non rational parameters, and discusses the relationship between controlling parameters, weighted factors and types, kinds and characteristics of curve segments and curved surface fragments. A mathematical method is provided for CAGD with abundant connotations, broad covering region, convenience, flexibility and direct simplicity.展开更多
Pulmonary vessels extraction is a challenging task in clinical medicine. Many pulmonary diseases are accompanied by the changes of vessel diameters. The vessels and their branches, which exhibit much variability, are ...Pulmonary vessels extraction is a challenging task in clinical medicine. Many pulmonary diseases are accompanied by the changes of vessel diameters. The vessels and their branches, which exhibit much variability, are most important in performing diagnosis and planning the follow-up therapies. In this paper, we propose an efficient approach to pulmonary vessels extraction based on the curve evolution. This approach models the vessels as monotonically marching front under the speed field integrating both the region and the edge information where a new region speed function is designed and integrated with the edge based speed function. Due to the region based speed term, the front could even propagate in small narrow vessel branches. To further improve the segmentation results, a multi-initial fast marching algorithm is developed to fast implement the numerical solution, which may avoid the monotonically marching front leaking out of the weak boundary too earlier and also reduce the computational cost. The validity of our approach is demonstrated by CT pulmonary vessels extraction. Experiments show that the segmentation results by our approach, especially on the narrow thin vessel branches extraction, are more precise than that of the existing method.展开更多
This paper discusses the problem that constructing a curve to satisfy the given endpoint constraints and chord-length parameters. Based on the research of Lu, the curve construction method for the entire tangent angle...This paper discusses the problem that constructing a curve to satisfy the given endpoint constraints and chord-length parameters. Based on the research of Lu, the curve construction method for the entire tangent angles region （α0, α1）∈（-r, r）×（-r, r） is given. Firstly, to ensure the weights are always positive, the three characteristics of cubic rational Bezier curve is proved, then the segment construction idea for the other tangent angles are presented in view of the three characteristics. The curve constructed with the new method satisfies the endpoint constraint and chord-length parameters, it＇s G1 continuous in every segment curve, and the shapes of the curve are well.展开更多
Well logging curves serve as indicators of strata attribute changes and are frequently utilized for stratigraphic analysis and comparison.Deep learning,known for its robust feature extraction capabilities,has seen con...Well logging curves serve as indicators of strata attribute changes and are frequently utilized for stratigraphic analysis and comparison.Deep learning,known for its robust feature extraction capabilities,has seen continuous adoption by scholars in the realm of well logging stratigraphic correlation tasks.Nonetheless,current deep learning algorithms often struggle to accurately capture feature changes occurring at layer boundaries within the curves.Moreover,when faced with data imbalance issues,neural networks encounter challenges in accurately modeling the one-hot encoded curve stratifi cation positions,resulting in signifi cant deviations between predicted and actual stratifi cation positions.Addressing these challenges,this study proposes a novel well logging curve stratigraphic comparison algorithm based on uniformly distributed soft labels.In the training phase,a label smoothing loss function is introduced to comprehensively account for the substantial loss stemming from data imbalance and to consider the similarity between diff erent layer data.Concurrently,spatial attention and channel attention mechanisms are incorporated into the shallow and deep encoder stages of U²-Net,respectively,to better focus on changes in stratifi cation positions.During the prediction phase,an optimized confi dence threshold algorithm is proposed to constrain stratifi cation results and solve the problem of reduced prediction accuracy because of occasional layer repetition.The proposed method is applied to real-world well logging data in oil fi elds.Quantitative evaluation results demonstrate that within error ranges of 1,2,and 3 m,the accuracy of well logging curve stratigraphic division reaches 87.27%,92.68%,and 95.08%,respectively,thus validating the eff ectiveness of the algorithm presented in this paper.展开更多
To reduce the difficulty of implementation and shorten the runtime of the traditional Kim-Fisher model, an entirely discrete Kim-Fisher-like model on lattices is proposed. The discrete model is directly built on the l...To reduce the difficulty of implementation and shorten the runtime of the traditional Kim-Fisher model, an entirely discrete Kim-Fisher-like model on lattices is proposed. The discrete model is directly built on the lattices, and the greedy algorithm is used in the implementation to continually decrease the energy function. First, regarding the gray values in images as discrete-valued random variables makes it possible to make a much simpler estimation of conditional entropy. Secondly, a uniform method within the level set framework for two-phase and multiphase segmentations without extension is presented. Finally, a more accurate approximation to the curve length on lattices with multi-labels is proposed. The experimental results show that, compared with the continuous Kim-Fisher model, the proposed model can obtain comparative results, while the implementation is much simpler and the runtime is dramatically reduced.展开更多
Level Set methods are robust and efficient numerical tools for resolving curve evolution in image segmentation. This paper proposes a new image segmentation algorithm based on Mumford-Shah module. The method is used t...Level Set methods are robust and efficient numerical tools for resolving curve evolution in image segmentation. This paper proposes a new image segmentation algorithm based on Mumford-Shah module. The method is used to CT images and the experiment results demonstrate its efficiency and veracity.展开更多
光学字符识别(Optical Character Recognition,OCR)是对文本图片进行扫描,然后对图像进行分析处理,获取到其中的文字内容的过程。但是目前的OCR算法对于弯曲的长文本普遍识别效果不佳,为此,提出了一种面向识别的长弯曲文本预处理算法,...光学字符识别(Optical Character Recognition,OCR)是对文本图片进行扫描,然后对图像进行分析处理,获取到其中的文字内容的过程。但是目前的OCR算法对于弯曲的长文本普遍识别效果不佳,为此,提出了一种面向识别的长弯曲文本预处理算法,即在文本行识别之前添加长弯曲文本处理模块(Long Curve Text Processing,LCTP),以提升图像中所有文本行识别的准确率。首先,在进行文本区域检测后,获取单条长弯曲文本行并清除干扰信息;其次,根据单条长弯曲文本行的特征计算每条弯曲文本行的关键拐点;进而,使用关键拐点对单条文本行进行切分和融合;最后,将经过切分与融合后的文本行输入文本行识别模型中得到最终识别结果。通过手动采集长弯曲文本图像形成的数据集Long Curve Text与目前主流OCR框架PP-OCR和Tesseract OCR进行对比实验可知,LA、MED、NED指标均有提升,相比于PP-OCR,LA提升49.5%,MED和NED分别降低了44115和0.182;相比于Tesseract OCR,LA提升3.2%,MED和NED分别降低了30282和0.125。同时,也在Long Curve Text数据集中进行了消融实验以验证本文提出LCTP的有效性以及进行了LCTP各个结构的时间对比实验以验证本文提出LCTP的高效性。结果表明LCTP可以提高长弯曲文本识别准确率,总体上可以地获得更加准确、有效的识别结果。展开更多
基金Project Supported by the National Natural Science Foundation of China (No.49801016 and 49671063)
文摘In this paper, the positional error curve of point features was extended to an error curves band of line segment features. Firstly, the constitution and shape of the error curves band were analyzed. On this basis, the general boundary curve formula of that band was derived. Secondly, the visualizing error curves bands were realized through three exam- ples. Finally,area index has been examined by comparing numerical results from error curves band and error ellipes band.
文摘By adopting the method of controlling parameters this paper describes the construction of various kinds of cubic curve segment and curved surface fragment with rational and non rational parameters, and discusses the relationship between controlling parameters, weighted factors and types, kinds and characteristics of curve segments and curved surface fragments. A mathematical method is provided for CAGD with abundant connotations, broad covering region, convenience, flexibility and direct simplicity.
基金Supported by the national Natural Science Foundation of China under Grant No.6 0 2 710 2 2 and the Creative Research Group Science Foundation of China under Grant No.6 0 0 2 4 30 1
文摘Pulmonary vessels extraction is a challenging task in clinical medicine. Many pulmonary diseases are accompanied by the changes of vessel diameters. The vessels and their branches, which exhibit much variability, are most important in performing diagnosis and planning the follow-up therapies. In this paper, we propose an efficient approach to pulmonary vessels extraction based on the curve evolution. This approach models the vessels as monotonically marching front under the speed field integrating both the region and the edge information where a new region speed function is designed and integrated with the edge based speed function. Due to the region based speed term, the front could even propagate in small narrow vessel branches. To further improve the segmentation results, a multi-initial fast marching algorithm is developed to fast implement the numerical solution, which may avoid the monotonically marching front leaking out of the weak boundary too earlier and also reduce the computational cost. The validity of our approach is demonstrated by CT pulmonary vessels extraction. Experiments show that the segmentation results by our approach, especially on the narrow thin vessel branches extraction, are more precise than that of the existing method.
基金Supported by Shandong Province Higher Educational Science and Technology Program(No.J12LN34)Shandong Ji'nan College and Institute Independent Innovation Project(No.201303011,No.201303021,No.201303016)
文摘This paper discusses the problem that constructing a curve to satisfy the given endpoint constraints and chord-length parameters. Based on the research of Lu, the curve construction method for the entire tangent angles region （α0, α1）∈（-r, r）×（-r, r） is given. Firstly, to ensure the weights are always positive, the three characteristics of cubic rational Bezier curve is proved, then the segment construction idea for the other tangent angles are presented in view of the three characteristics. The curve constructed with the new method satisfies the endpoint constraint and chord-length parameters, it＇s G1 continuous in every segment curve, and the shapes of the curve are well.
基金supported by the CNPC Advanced Fundamental Research Projects(No.2023ycq06).
文摘Well logging curves serve as indicators of strata attribute changes and are frequently utilized for stratigraphic analysis and comparison.Deep learning,known for its robust feature extraction capabilities,has seen continuous adoption by scholars in the realm of well logging stratigraphic correlation tasks.Nonetheless,current deep learning algorithms often struggle to accurately capture feature changes occurring at layer boundaries within the curves.Moreover,when faced with data imbalance issues,neural networks encounter challenges in accurately modeling the one-hot encoded curve stratifi cation positions,resulting in signifi cant deviations between predicted and actual stratifi cation positions.Addressing these challenges,this study proposes a novel well logging curve stratigraphic comparison algorithm based on uniformly distributed soft labels.In the training phase,a label smoothing loss function is introduced to comprehensively account for the substantial loss stemming from data imbalance and to consider the similarity between diff erent layer data.Concurrently,spatial attention and channel attention mechanisms are incorporated into the shallow and deep encoder stages of U²-Net,respectively,to better focus on changes in stratifi cation positions.During the prediction phase,an optimized confi dence threshold algorithm is proposed to constrain stratifi cation results and solve the problem of reduced prediction accuracy because of occasional layer repetition.The proposed method is applied to real-world well logging data in oil fi elds.Quantitative evaluation results demonstrate that within error ranges of 1,2,and 3 m,the accuracy of well logging curve stratigraphic division reaches 87.27%,92.68%,and 95.08%,respectively,thus validating the eff ectiveness of the algorithm presented in this paper.
文摘To reduce the difficulty of implementation and shorten the runtime of the traditional Kim-Fisher model, an entirely discrete Kim-Fisher-like model on lattices is proposed. The discrete model is directly built on the lattices, and the greedy algorithm is used in the implementation to continually decrease the energy function. First, regarding the gray values in images as discrete-valued random variables makes it possible to make a much simpler estimation of conditional entropy. Secondly, a uniform method within the level set framework for two-phase and multiphase segmentations without extension is presented. Finally, a more accurate approximation to the curve length on lattices with multi-labels is proposed. The experimental results show that, compared with the continuous Kim-Fisher model, the proposed model can obtain comparative results, while the implementation is much simpler and the runtime is dramatically reduced.
文摘Level Set methods are robust and efficient numerical tools for resolving curve evolution in image segmentation. This paper proposes a new image segmentation algorithm based on Mumford-Shah module. The method is used to CT images and the experiment results demonstrate its efficiency and veracity.
