The paper presents a concept and implementation of an algorithm whose task is to automatically determine the outline of a spot representing cancerous tissue. We assume that the location of this tissue was previously d...
详细信息
The paper presents a concept and implementation of an algorithm whose task is to automatically determine the outline of a spot representing cancerous tissue. We assume that the location of this tissue was previously determined based on data obtained during computed tomography (CT). In medical images obtained on the basis of CT, active areas are represented by spots of appropriate intensity. For the purposes of properly selected radiotherapy, it is necessary to determine the outline of these areas. At the same time, it is very important to maintain an appropriate margin around the main area where the cancer occurs, so as not to miss small cancer foci located near the larger area. The aim of the proposed method is to process the biomedical image in such a way that the mentioned small foci are not lost, but are located inside the outline designated with an appropriate margin. The algorithm is based on a modified wavelet transform and appropriately selected linear and nonlinear filters with the possibility of their returning. The algorithm was, as a prototype, implemented in the Matlab environment and verified on images selected so that they were representative for various situations. The system for which the proposed algorithm is being developed operates in real time. For this reason, we paid special attention to the problem of computationalcomplexity of its particular stages. This is also important because during a medical procedure we deal with a series of biomedical images, creating a 3-D structure. Savings in this area obtained for a single image therefore give measurable benefits for the entire set of images.
In this paper, we introduce a novel Binary Directional Shape (BDS) descriptor for efficient object shape classification. The BDS descriptor is designed to operate in real-time environments, balancing computational eff...
详细信息
In this paper, we introduce a novel Binary Directional Shape (BDS) descriptor for efficient object shape classification. The BDS descriptor is designed to operate in real-time environments, balancing computational efficiency with high classification accuracy. The descriptor works by resizing input images, thresholding them into binary representations, and analyzing them along multiple predefined directions. Configurable parameters such as image size, number of directions, and step size allow for customization based on application needs. Through extensive experiments, we explore the effects of image resolution on shape classification, demonstrating how aliasing and resolution loss can impact performance. Our results show that while higher resolutions improve accuracy, challenges such as rotation variance and aliasing persist. The BDS descriptor outperforms traditional feature-based methods, providing superior classification results even in the presence of these challenges. This work demonstrates the value of directional information in shape recognition and sets the stage for future optimizations and applications.
This study proposes a novel model predictive control (MPC) based on receding horizon particle swarm optimisation (RHPSO) for formation control of non-holonomic mobile robots by incorporating collision avoidance and co...
详细信息
This study proposes a novel model predictive control (MPC) based on receding horizon particle swarm optimisation (RHPSO) for formation control of non-holonomic mobile robots by incorporating collision avoidance and control input minimisation and guaranteeing asymptotic stability. In most conventional MPC approaches, the collision avoidance constraint is imposed by the 2-norm of a relative position vector at each discrete time step. Thus, multi-robot formation control problem can be formulated as a constrained non-linear optimisation problem. In general, traditional optimisation techniques suitable for addressing constrained non-linear optimisation problems take a longer computation time with an increase in the number of constraints. The traditional approaches therefore suffer from the computational complexity problem corresponding to an increase in the prediction horizon. To address this problem without a significant increase in computationalcomplexity, a novel strategy for collision avoidance is proposed to incorporating a particle swarm optimisation. In addition, the stability conditions are derived in simplified forms that can be satisfied by selecting appropriate constant values for control gains and weight parameters. Numerical simulations verify the effectiveness of the proposed RHPSO-based formation control.
暂无评论