Obtaining a 3D feature description with high descriptiveness and robustness under complicated nuisances is a significant and challenging task in 3D feature matching. This paper proposes a novel feature description consisting of a stable local reference frame (LRF) and a feature descriptor based on local spatial voxels. First, an improved LRF was designed by incorporating distance weights into Z- and X-axis calculations. Subsequently, based on the LRF and voxel segmentation, a feature descriptor based on voxel homogenization was proposed. Moreover, uniform segmentation of cube voxels was performed, considering the eigenvalues of each voxel and its neighboring voxels, thereby enhancing the stability of the description. The performance of the descriptor was strictly tested and evaluated on three public datasets, which exhibited high descriptiveness, robustness, and superior performance compared with other current methods. Furthermore, the descriptor was applied to a 3D registration trial, and the results demonstrated the reliability of our approach.
{"title":"Novel 3D local feature descriptor of point clouds based on spatial voxel homogenization for feature matching.","authors":"Jiong Yang, Jian Zhang, Zhengyang Cai, Dongyang Fang","doi":"10.1186/s42492-023-00145-4","DOIUrl":"10.1186/s42492-023-00145-4","url":null,"abstract":"<p><p>Obtaining a 3D feature description with high descriptiveness and robustness under complicated nuisances is a significant and challenging task in 3D feature matching. This paper proposes a novel feature description consisting of a stable local reference frame (LRF) and a feature descriptor based on local spatial voxels. First, an improved LRF was designed by incorporating distance weights into Z- and X-axis calculations. Subsequently, based on the LRF and voxel segmentation, a feature descriptor based on voxel homogenization was proposed. Moreover, uniform segmentation of cube voxels was performed, considering the eigenvalues of each voxel and its neighboring voxels, thereby enhancing the stability of the description. The performance of the descriptor was strictly tested and evaluated on three public datasets, which exhibited high descriptiveness, robustness, and superior performance compared with other current methods. Furthermore, the descriptor was applied to a 3D registration trial, and the results demonstrated the reliability of our approach.</p>","PeriodicalId":29931,"journal":{"name":"Visual Computing for Industry Biomedicine and Art","volume":null,"pages":null},"PeriodicalIF":2.8,"publicationDate":"2023-09-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10539243/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41177182","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-06-14DOI: 10.1186/s42492-022-00108-1
Cao, Weiguo, Pomeroy, Marc J., Liang, Zhengrong, Abbasi, Almas F., Pickhardt, Perry J., Lu, Hongbing
Textures have become widely adopted as an essential tool for lesion detection and classification through analysis of the lesion heterogeneities. In this study, higher order derivative images are being employed to combat the challenge of the poor contrast across similar tissue types among certain imaging modalities. To make good use of the derivative information, a novel concept of vector texture is firstly introduced to construct and extract several types of polyp descriptors. Two widely used differential operators, i.e., the gradient operator and Hessian operator, are utilized to generate the first and second order derivative images. These derivative volumetric images are used to produce two angle-based and two vector-based (including both angle and magnitude) textures. Next, a vector-based co-occurrence matrix is proposed to extract texture features which are fed to a random forest classifier to perform polyp classifications. To evaluate the performance of our method, experiments are implemented over a private colorectal polyp dataset obtained from computed tomographic colonography. We compare our method with four existing state-of-the-art methods and find that our method can outperform those competing methods over 4%-13% evaluated by the area under the receiver operating characteristics curves.
{"title":"Vector textures derived from higher order derivative domains for classification of colorectal polyps","authors":"Cao, Weiguo, Pomeroy, Marc J., Liang, Zhengrong, Abbasi, Almas F., Pickhardt, Perry J., Lu, Hongbing","doi":"10.1186/s42492-022-00108-1","DOIUrl":"https://doi.org/10.1186/s42492-022-00108-1","url":null,"abstract":"Textures have become widely adopted as an essential tool for lesion detection and classification through analysis of the lesion heterogeneities. In this study, higher order derivative images are being employed to combat the challenge of the poor contrast across similar tissue types among certain imaging modalities. To make good use of the derivative information, a novel concept of vector texture is firstly introduced to construct and extract several types of polyp descriptors. Two widely used differential operators, i.e., the gradient operator and Hessian operator, are utilized to generate the first and second order derivative images. These derivative volumetric images are used to produce two angle-based and two vector-based (including both angle and magnitude) textures. Next, a vector-based co-occurrence matrix is proposed to extract texture features which are fed to a random forest classifier to perform polyp classifications. To evaluate the performance of our method, experiments are implemented over a private colorectal polyp dataset obtained from computed tomographic colonography. We compare our method with four existing state-of-the-art methods and find that our method can outperform those competing methods over 4%-13% evaluated by the area under the receiver operating characteristics curves.","PeriodicalId":29931,"journal":{"name":"Visual Computing for Industry Biomedicine and Art","volume":null,"pages":null},"PeriodicalIF":2.8,"publicationDate":"2022-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138495077","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Deep simulations have gained widespread attention owing to their excellent acceleration performances. However, these methods cannot provide effective collision detection and response strategies. We propose a deep interactive physical simulation framework that can effectively address tool-object collisions. The framework can predict the dynamic information by considering the collision state. In particular, the graph neural network is chosen as the base model, and a collision-aware recursive regression module is introduced to update the network parameters recursively using interpenetration distances calculated from the vertex-face and edge-edge tests. Additionally, a novel self-supervised collision term is introduced to provide a more compact collision response. This study extensively evaluates the proposed method and shows that it effectively reduces interpenetration artifacts while ensuring high simulation efficiency.
{"title":"Collision-aware interactive simulation using graph neural networks","authors":"Zhu, Xin, Qian, Yinling, Wang, Qiong, Feng, Ziliang, Heng, Pheng-Ann","doi":"10.1186/s42492-022-00113-4","DOIUrl":"https://doi.org/10.1186/s42492-022-00113-4","url":null,"abstract":"Deep simulations have gained widespread attention owing to their excellent acceleration performances. However, these methods cannot provide effective collision detection and response strategies. We propose a deep interactive physical simulation framework that can effectively address tool-object collisions. The framework can predict the dynamic information by considering the collision state. In particular, the graph neural network is chosen as the base model, and a collision-aware recursive regression module is introduced to update the network parameters recursively using interpenetration distances calculated from the vertex-face and edge-edge tests. Additionally, a novel self-supervised collision term is introduced to provide a more compact collision response. This study extensively evaluates the proposed method and shows that it effectively reduces interpenetration artifacts while ensuring high simulation efficiency.","PeriodicalId":29931,"journal":{"name":"Visual Computing for Industry Biomedicine and Art","volume":null,"pages":null},"PeriodicalIF":2.8,"publicationDate":"2022-06-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138495076","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Facial expression recognition (FER) has numerous applications in computer security, neuroscience, psychology, and engineering. Owing to its non-intrusiveness, it is considered a useful technology for combating crime. However, FER is plagued with several challenges, the most serious of which is its poor prediction accuracy in severe head poses. The aim of this study, therefore, is to improve the recognition accuracy in severe head poses by proposing a robust 3D head-tracking algorithm based on an ellipsoidal model, advanced ensemble of AdaBoost, and saturated vector machine (SVM). The FER features are tracked from one frame to the next using the ellipsoidal tracking model, and the visible expressive facial key points are extracted using Gabor filters. The ensemble algorithm (Ada-AdaSVM) is then used for feature selection and classification. The proposed technique is evaluated using the Bosphorus, BU-3DFE, MMI, CK + , and BP4D-Spontaneous facial expression databases. The overall performance is outstanding.
面部表情识别(FER)在计算机安全、神经科学、心理学和工程学中有着广泛的应用。由于其非侵入性，它被认为是打击犯罪的有用技术。然而，该方法面临着一些挑战，其中最严重的是对严重头部姿势的预测精度较差。为此，本研究提出了一种基于椭球体模型、AdaBoost高级集成和饱和向量机(SVM)的鲁棒三维头部跟踪算法，以提高严重头部姿态的识别精度。利用椭球体跟踪模型对图像进行逐帧跟踪，并利用Gabor滤波器提取可见的面部表情关键点。然后使用集成算法(Ada-AdaSVM)进行特征选择和分类。使用Bosphorus, BU-3DFE, MMI, CK +和bp4d -自发面部表情数据库对所提出的技术进行评估。整体表现非常出色。
{"title":"Robust facial expression recognition system in higher poses","authors":"Owusu, Ebenezer, Appati, Justice Kwame, Okae, Percy","doi":"10.1186/s42492-022-00109-0","DOIUrl":"https://doi.org/10.1186/s42492-022-00109-0","url":null,"abstract":"Facial expression recognition (FER) has numerous applications in computer security, neuroscience, psychology, and engineering. Owing to its non-intrusiveness, it is considered a useful technology for combating crime. However, FER is plagued with several challenges, the most serious of which is its poor prediction accuracy in severe head poses. The aim of this study, therefore, is to improve the recognition accuracy in severe head poses by proposing a robust 3D head-tracking algorithm based on an ellipsoidal model, advanced ensemble of AdaBoost, and saturated vector machine (SVM). The FER features are tracked from one frame to the next using the ellipsoidal tracking model, and the visible expressive facial key points are extracted using Gabor filters. The ensemble algorithm (Ada-AdaSVM) is then used for feature selection and classification. The proposed technique is evaluated using the Bosphorus, BU-3DFE, MMI, CK + , and BP4D-Spontaneous facial expression databases. The overall performance is outstanding.","PeriodicalId":29931,"journal":{"name":"Visual Computing for Industry Biomedicine and Art","volume":null,"pages":null},"PeriodicalIF":2.8,"publicationDate":"2022-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138495075","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-05-10DOI: 10.1186/s42492-022-00111-6
S. Amraee, Maryam Chinipardaz, Mohammadali Charoosaei
{"title":"Analytical study of two feature extraction methods in comparison with deep learning methods for classification of small metal objects","authors":"S. Amraee, Maryam Chinipardaz, Mohammadali Charoosaei","doi":"10.1186/s42492-022-00111-6","DOIUrl":"https://doi.org/10.1186/s42492-022-00111-6","url":null,"abstract":"","PeriodicalId":29931,"journal":{"name":"Visual Computing for Industry Biomedicine and Art","volume":null,"pages":null},"PeriodicalIF":2.8,"publicationDate":"2022-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"65794142","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-04-08DOI: 10.1186/s42492-022-00107-2
Ballambat, Raghuvir Pai, Zuber, Mohammad, Khader, Shah Mohammed Abdul, Ayachit, Anurag, Ahmad, Kamarul Arifin bin, Vedula, Rajanikanth Rao, Kamath, Sevagur Ganesh, Shuaib, Ibrahim Lutfi
Cerebral intracranial aneurysms are serious problems that can lead to stroke, coma, and even death. The effect of blood flow on cerebral aneurysms and their relationship with rupture are unknown. In addition, postural changes and their relevance to haemodynamics of blood flow are difficult to measure in vivo using clinical imaging alone. Computational simulations investigating the detailed haemodynamics in cerebral aneurysms have been developed in recent times not only to understand the progression and rupture but also for clinical evaluation and treatment. In the present study, the haemodynamics of a patient-specific case of a large aneurysm on the left side internal carotid bifurcation (LICA) and no aneurysm on the right side internal carotid bifurcation (RICA) was investigated. The simulation of these patient-specific models using fluid–structure interaction provides a valuable comparison of flow behavior between normal and aneurysm models. The influences of postural changes were investigated during standing, sleeping, and head-down (HD) position. Significant changes in flow were observed during the HD position and quit high arterial blood pressure in the internal carotid artery (ICA) aneurysm model was established when compared to the normal ICA model. The velocity increased abruptly during the HD position by more than four times (LICA and RICA) and wall shear stress by four times (LICA) to ten times (RICA). The complex spiral flow and higher pressures prevailing within the dome increase the risk of aneurysm rupture.
{"title":"Influence of postural changes on haemodynamics in internal carotid artery bifurcation aneurysm using numerical methods","authors":"Ballambat, Raghuvir Pai, Zuber, Mohammad, Khader, Shah Mohammed Abdul, Ayachit, Anurag, Ahmad, Kamarul Arifin bin, Vedula, Rajanikanth Rao, Kamath, Sevagur Ganesh, Shuaib, Ibrahim Lutfi","doi":"10.1186/s42492-022-00107-2","DOIUrl":"https://doi.org/10.1186/s42492-022-00107-2","url":null,"abstract":"Cerebral intracranial aneurysms are serious problems that can lead to stroke, coma, and even death. The effect of blood flow on cerebral aneurysms and their relationship with rupture are unknown. In addition, postural changes and their relevance to haemodynamics of blood flow are difficult to measure in vivo using clinical imaging alone. Computational simulations investigating the detailed haemodynamics in cerebral aneurysms have been developed in recent times not only to understand the progression and rupture but also for clinical evaluation and treatment. In the present study, the haemodynamics of a patient-specific case of a large aneurysm on the left side internal carotid bifurcation (LICA) and no aneurysm on the right side internal carotid bifurcation (RICA) was investigated. The simulation of these patient-specific models using fluid–structure interaction provides a valuable comparison of flow behavior between normal and aneurysm models. The influences of postural changes were investigated during standing, sleeping, and head-down (HD) position. Significant changes in flow were observed during the HD position and quit high arterial blood pressure in the internal carotid artery (ICA) aneurysm model was established when compared to the normal ICA model. The velocity increased abruptly during the HD position by more than four times (LICA and RICA) and wall shear stress by four times (LICA) to ten times (RICA). The complex spiral flow and higher pressures prevailing within the dome increase the risk of aneurysm rupture.","PeriodicalId":29931,"journal":{"name":"Visual Computing for Industry Biomedicine and Art","volume":null,"pages":null},"PeriodicalIF":2.8,"publicationDate":"2022-04-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138495074","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-04-01DOI: 10.1186/s42492-022-00106-3
Cindy Xue, J. Yuan, Yihang Zhou, O. Wong, K. Cheung, S. Yu
{"title":"Acquisition repeatability of MRI radiomics features in the head and neck: a dual-3D-sequence multi-scan study","authors":"Cindy Xue, J. Yuan, Yihang Zhou, O. Wong, K. Cheung, S. Yu","doi":"10.1186/s42492-022-00106-3","DOIUrl":"https://doi.org/10.1186/s42492-022-00106-3","url":null,"abstract":"","PeriodicalId":29931,"journal":{"name":"Visual Computing for Industry Biomedicine and Art","volume":null,"pages":null},"PeriodicalIF":2.8,"publicationDate":"2022-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"65794084","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-03-07DOI: 10.1186/s42492-022-00104-5
R. Cattell, Jia Ying, Lan Lei, Jie Ding, Shenglan Chen, Mario Serrano Sosa, Chuan Huang
{"title":"Preoperative prediction of lymph node metastasis using deep learning-based features","authors":"R. Cattell, Jia Ying, Lan Lei, Jie Ding, Shenglan Chen, Mario Serrano Sosa, Chuan Huang","doi":"10.1186/s42492-022-00104-5","DOIUrl":"https://doi.org/10.1186/s42492-022-00104-5","url":null,"abstract":"","PeriodicalId":29931,"journal":{"name":"Visual Computing for Industry Biomedicine and Art","volume":null,"pages":null},"PeriodicalIF":2.8,"publicationDate":"2022-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41576804","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号