KEYWORDS: Tumors, Liver, Diagnostics, Matrices, Computed tomography, Data modeling, Probability theory, Positron emission tomography, Diseases and disorders, Computing systems
Incomplete multi-view clustering (IMVC) for liver tumor CT images provide common and complementary information, that essential to support liver tumor diagnosis of without labeling. These IMVC techniques still have two problems, though: 1) they only heed the following to the resemblance between the same samples of different views, ignoring the resemblance between different samples of the same view, which can provide more information for liver tumor diagnose; 2) most of the existing diagnostic processes are staged, greatly increasing the cost of time. We suggest a solution to these problems by a dual contrastive learning framework to learning mutual information from cross-view contrast learning and intra-view contrast learning at the same time. On this basis, the cosine similarity between views is constructed using positive and negative samples from these dual contrastive learning, and the whole process is carried out synchronously. In this paper, two multi-view liver tumor data sets are compared with 4 methods in recent years, indicating that this method is the most advanced.
Large ships have a large deck area, many on-board equipment and devices require attitude information. if every device that requires attitude information is equipped with an attitude reference, Firstly, it will cause a waste of resources. Secondly, due to the influence of the deformation of large ships, the attitude reference of each device will be inconsistent, and then the collaborative work of large ships will be affected. Based on the attitude transfer theory, this paper uses genetic algorithm to optimize the layout of the attitude benchmarks of large ships according to the requirements of the shipboard equipment for attitude information. Simulation calculations show that this method can effectively reduce the number of benchmark layouts.
Object detection based on deep learning algorithms has been an important yet challenging research field in computer vision. The feature pyramid network has become a dominant network architecture in many detection applications because of its powerful feature learning ability for objects with varying scales. To address the challenges in detecting small and densely packed objects, this paper proposes an innovative object detection approach by combining the path aggregation scheme and the feature pyramid network into a unified framework. Specifically, we add a bottom-up branch with lateral connection onto the existing feature pyramid network and apply adaptive feature fusion strategy, which improves the detection performance for small and densely arranged objects in remote sensing images. Experiment results show that our proposed path aggregated feature pyramid network can improve the detection performance for diverse objects in aerial images.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.