-
Chapter and Conference Paper
Shadow Detection of Remote Sensing Image by Fusion of Involution and Shunted Transformer
In the field of remote sensing image analysis, this paper addresses the challenges of accurately detecting edge details and small shadow regions through the introduction of STAISD-Net. This end-to-end network ...
-
Chapter and Conference Paper
Load-Path Method for Design of Disturbed Regions in Concrete Structures
This study develops a quantitative load path method for the shape selection of the strut-and-tie models in disturbed regions (D-regions) of structural concrete. Firstly, quantitative load paths are generated i...
-
Chapter and Conference Paper
State Estimation of Batch Processes Based on Data Synchronization Technology Under Unknown Measurement Delay
Batch processes are important production processes that have been widely used in industry. However, some important quality variables are difficult to accurately measure due to cost or hardware constraints. The...
-
Chapter and Conference Paper
Hybrid Soft Sensor Modeling of Batch Processes Based on Risk Assessment
Batch processes are time-varying, high-dimensional, and highly nonlinear. Some primary variables are difficult to be measured online in real-time. Soft sensor modeling plays a critical role in estimating the p...
-
Chapter and Conference Paper
Improving Adversarial Robustness of Deep Neural Networks via Linear Programming
Adversarial training provides an effective means to improve the robustness of neural networks against adversarial attacks. The nonlinear feature of neural networks makes it difficult to find good adversarial e...
-
Chapter and Conference Paper
Multi-scale Superpixel Based Homomorphic Filtering for the Component Image Enhancement
Aiming at the low accuracy of visual measurement caused by the nonuniform illumination image of components during industrial processing and assembly, a multi-scale superpixel based homomorphic filtering method...
-
Chapter and Conference Paper
Multi-scale Visual Saliency Fusing with Spatiotemporal Features for Fire Detection
Aiming at the problem that the existing methods cannot realize the fire detection accurately, a multi-scale visual saliency fusing with spatiotemporal features for fire detection method is proposed. Firstly, t...
-
Chapter and Conference Paper
Specular Reflection Separation for a Single Image Using IR-SIP
Specular reflection separation can improve the accuracy of edge detection, target recognition, and feature matching. Existing methods can’t completely separate the specular component for a single image. We pro...
-
Chapter and Conference Paper
A Heilbronn Type Inequality for Plane Nonagons
In this paper, we present a proof of the property that for any convex nonagon \(P_1P_2\ldots P_9\) in the plane, the smallest area of a triangle \(P_{i}P_{j}P_{k} (1\le i< j < k \le 9)\) is at most a fraction of
-
Chapter and Conference Paper
WIFI-Based Indoor Positioning System with Twice Clustering and Multi-user Topology Approximation Algorithm
In recent years, indoor positioning technology based on WIFI has been widely researched. However, traditional WIFI-based indoor positioning method can’t achieve high localization accuracy due to the clustering...