Object Detection and Tracking with YOLO and the Sliding Innovation Filter.

Object detection and tracking are pivotal tasks in machine learning, particularly within the domain of computer vision technologies. Despite significant advancements in object detection frameworks, challenges persist in real-world tracking scenarios, including object interactions, occlusions, and background interference. Many algorithms have been proposed to carry out such tasks; however, most struggle to perform well in the face of disturbances and uncertain environments. This research proposes a novel approach by integrating the You Only Look Once (YOLO) architecture for object detection with a robust filter for target tracking, addressing issues of disturbances and uncertainties. The YOLO architecture, known for its real-time object detection capabilities, is employed for initial object detection and centroid location. In combination with the detection framework, the sliding innovation filter, a novel robust filter, is implemented and postulated to improve tracking reliability in the face of disturbances. Specifically, the sliding innovation filter is implemented to enhance tracking performance by estimating the optimal centroid location in each frame and updating the object's trajectory. Target tracking traditionally relies on estimation theory techniques like the Kalman filter, and the sliding innovation filter is introduced as a robust alternative particularly suitable for scenarios where a priori information about system dynamics and noise is limited. Experimental simulations in a surveillance scenario demonstrate that the sliding innovation filter-based tracking approach outperforms existing Kalman-based methods, especially in the presence of disturbances. In all, this research contributes a practical and effective approach to object detection and tracking, addressing challenges in real-world, dynamic environments. The comparative analysis with traditional filters provides practical insights, laying the groundwork for future work aimed at advancing multi-object detection and tracking capabilities in diverse applications.

Interacting Multiple Model Estimators for Fault Detection in a Magnetorheological Damper.

This paper proposes a novel estimator for the purpose of fault detection and diagnosis. The interacting multiple model (IMM) strategy is effective for estimating the behaviour of systems with multiple operating modes. Each mode corresponds to a distinct mathematical model and is subject to a filtering process. This paper applies various model-based filters in combination with the IMM strategy. One such estimator employs the recently introduced extended sliding innovation filter (ESIF) known as the IMM-ESIF. The ESIF is an extension of the sliding innovation filter for nonlinear systems based on the sliding mode concept. In the presence of modeling uncertainties, the ESIF has been proven to be more robust compared to methods such as the extended Kalman filter (EKF). The novel IMM-ESIF strategy is also compared with the IMM strategy, which incorporates the unscented Kalman filter (UKF), referred to herein as IMM-UKF. While EKF uses Taylor series approximation to linearize the system model, the UKF uses sigma point to calculate the system's mean and covariance. The methods were applied to an experimental magnetorheological (MR) damper setup, which was designed for testing control and estimation theory. Magnetorheological dampers exhibit a diverse array of applications in the automotive and aerospace sectors, with particular relevance to attenuating vibrations through adaptive suspension systems. Applied to a magnetorheological (MR) damper with distinct operating modes determined by the damper's current, the results showcase the effectiveness of IMM-ESIF. In mixed operational conditions, IMM-ESIF demonstrates a notable 80% to 90% reduction in estimation error compared to its counterparts. Furthermore, it exhibits a 4% to 5% enhancement in correctly classifying operational modes, establishing IMM-ESIF as a promising and efficient alternative for adaptive estimation in electromechanical systems. The improved accuracy in estimating the system's behaviour, even amidst uncertainties and mixed operational scenarios, signifies the potential of IMM-ESIF to significantly enhance the overall robustness and efficiency of estimations.

Machine learning algorithms in microbial classification: a comparative analysis.

This research paper presents an overview of contemporary machine learning methodologies and their utilization in the domain of healthcare and the prevention of infectious diseases, specifically focusing on the classification and identification of bacterial species. As deep learning techniques have gained prominence in the healthcare sector, a diverse array of architectural models has emerged. Through a comprehensive review of pertinent literature, multiple studies employing machine learning algorithms in the context of microbial diagnosis and classification are examined. Each investigation entails a tabulated presentation of data, encompassing details about the training and validation datasets, specifications of the machine learning and deep learning techniques employed, as well as the evaluation metrics utilized to gauge algorithmic performance. Notably, Convolutional Neural Networks have been the predominant selection for image classification tasks by machine learning practitioners over the last decade. This preference stems from their ability to autonomously extract pertinent and distinguishing features with minimal human intervention. A range of CNN architectures have been developed and effectively applied in the realm of image classification. However, addressing the considerable data requirements of deep learning, recent advancements encompass the application of pre-trained models using transfer learning for the identification of microbial entities. This method involves repurposing the knowledge gleaned from solving alternate image classification challenges to accurately classify microbial images. Consequently, the necessity for extensive and varied training data is significantly mitigated. This study undertakes a comparative assessment of various popular pre-trained CNN architectures for the classification of bacteria. The dataset employed is composed of approximately 660 images, representing 33 bacterial species. To enhance dataset diversity, data augmentation is implemented, followed by evaluation on multiple models including AlexNet, VGGNet, Inception networks, Residual Networks, and Densely Connected Convolutional Networks. The results indicate that the DenseNet-121 architecture yields the optimal performance, achieving a peak accuracy of 99.08%, precision of 99.06%, recall of 99.00%, and an F1-score of 98.99%. By demonstrating the proficiency of the DenseNet-121 model on a comparatively modest dataset, this study underscores the viability of transfer learning in the healthcare sector for precise and efficient microbial identification. These findings contribute to the ongoing endeavors aimed at harnessing machine learning techniques to enhance healthcare methodologies and bolster infectious disease prevention practices.

Zinc(ii)-based coordination polymer encapsulated Tb3+ as a multi-responsive luminescent sensor for Ru3+, Fe3+, CrO4 2-, Cr2O7 2- and MnO4.

A zinc(ii)-based coordination polymer (CP), namely [Zn(modbc)2] n (Zn-CP) (modbc = 2-methyl-6-oxygen-1,6-dihydro-3,4'-bipyridine-5-carbonitrile), has been synthesized and characterized. Single-crystal structural determination reveals that Zn-CP is a two-dimensional framework structure with tetranuclear homometallic Zn4(modbc)4 units cross-linked by modbc. The excellent luminescence as well as good stability of Zn-CP do not enable it to have selective sensing capability for different ions. After encapsulation of Tb3+ in Zn-CP, the as-obtained fluorescent functionalized Tb3+@Zn-CP maintained excellent luminescence as well as stability, which made it a highly selective and sensitive multiresponsive luminescent sensor for Ru3+, Fe3+, CrO4 2-, Cr2O7 2-, and MnO4 - with high sensitivity, good anti-interference performance, and quick response time (∼10 s). The detection limits are 0.27 µM, 0.57 µM, 0.10 µM, 0.43 µM and 0.15 µM, respectively. A possible sensing mechanism was discussed in detail.

Cu(ii)-TACN complexes selectively induce antitumor activity in HepG-2 cells via DNA damage and mitochondrial-ROS-mediated apoptosis.

A new 1,4,7-triazacyclononane derivative, 4-benzyloxy-benzyl-1,4,7-triazacyclononane (btacn), and three associated cyclen complexes, Cu(btacn)Cl2, Zn(btacn)Cl2 and [Cu(btacn)2]·(ClO4)2, were prepared to serve as DNA synthesis interferents. The compounds were characterized using IR, 1H and 13C NMR, ESI-MS, elemental analysis and X-ray single crystal diffraction methods, and their DNA damage mechanisms and cytotoxicities towards cancer and normal cells were studied. Among them, Cu(btacn)Cl2 and [Cu(btacn)2]·(ClO4)2 exhibit potent anti-proliferation activity in HepG-2 and HeLa cells, but low cytotoxicity in the normal cell models LO2 and HUVEC, giving SI values (IC50 ratios) ranging from 2.45 to 7.09-times higher than that of cisplatin. DNA binding and cleavage studies suggested that [Cu(btacn)2]·(ClO4)2 can more easily intercalate into CT-DNA than Cu(btacn)Cl2, which is consistent with the results of G2/S phase arrest and apoptosis in HepG-2 cells involving the complexes. In contrast, Zn(btacn)Cl2 demonstrated weak DNA binding and no obvious cytotoxicity. The results suggest that Cu(btacn)Cl2 and [Cu(btacn)2]·(ClO4)2 mainly undergo redox processes to produce reactive oxygen species (ROS) that induce DNA degradation and mitochondrial damage.

Two-dimensional Cd(ii) coordination polymer encapsulated by Tb3+ as a reversible luminescent probe for Fe3.

A two-dimensional luminescent cadmium(ii) coordination polymer, [Cd(modbc)2] n (Cd-P); modbc = 2-methyl-6-oxygen-1,6-dihydro-3,4'-bipyridine-5-carbonitrile, was successfully synthesized by a solvothermal reaction and fully characterized. Cd-P exhibited excellent luminescence emission, and detected Cu2+, Co2+, Fe2+, Hg2+, Ni2+ and Fe3+ ions with high sensitivity and showed good anti-interference performance. After encapsulation of Tb3+ ions in Cd-P, the as-obtained fluorescent functionalized Tb3+@Cd-P maintained distinct chemical stabilities in different pHs and metal salt solutions. Subsequently, we explored the potential application of Tb3+@Cd-P as a probe for Fe3+ ions. A new and convenient method for individual identification of Fe3+ ions by the combination of Cd-P and Tb3+@Cd-P was successfully established. A possible sensing mechanism is discussed in detail.