Spatial Tracking Research Articles

Sign Language Interpreting System Using Recursive Neural Networks

According to the World Health Organization (WHO), 5% of people around the world have hearing disabilities, which limits their capacity to communicate with others. Recently, scientists have proposed systems based on deep learning techniques to create a sign language-to-text translator, expecting this to help deaf people communicate; however, the performance of such systems is still low for practical scenarios. Furthermore, the proposed systems are language-oriented, which leads to particular problems related to the signs for each language. For this reason, to address this problem, in this paper, we propose a system based on a Recursive Neural Network (RNN) focused on Mexican Sign Language (MSL) that uses the spatial tracking of hands and facial expressions to predict the word that a person intends to communicate. To achieve this, we trained four RNN-based models using a dataset of 600 clips that were 30 s long; each word included 30 clips. We conducted two experiments; we tailored the first experiment to determine the most well-suited model for the target application and measure the accuracy of the resulting system in offline mode; in the second experiment, we measured the accuracy of the system in online mode. We assessed the system’s performance using the following metrics: the precision, recall, F1-score, and the number of errors during online scenarios, and the results computed indicate an accuracy of 0.93 in the offline mode and a higher performance for the online operating mode compared to previously proposed approaches. These results underscore the potential of the proposed scheme in scenarios such as teaching, learning, commercial transactions, and daily communications among deaf and non-deaf people.

Smart3DMOT: Smart cascade 3D MOT tracking strategy with motion and appearance association

3D multi-object tracking (MOT) is pivotal for associating the trajectories of objects in autonomous driving. The integration of image and point cloud data in 3D MOT algorithms has emerged as a research hotspot, aiming to achieve both high precision and efficiency tracking. In this paper, we present Smart3DMOT, an novel method that leverages multi-modal detection data and incorporates a cascade matching strategy, intelligently utilizing motion and multi-modal appearance affinity for data association across frames. Our algorithm introduces two association models: the strong motion association model (SM-AM) and the multi-modal fusion feature association model (MMFF-AM), tailored to different tracking scenarios. The SM-AM robustly associates objects by enhancing motion similarity, while the MMFF-AM integrates multi-modal fusion features from images and point clouds to generate appearance similarity between objects and combine motion cues for joint reasoning. Finally, we propose a cascade tracking strategy with a “spatial tracking state confirmer” (STSC). The SM-AM and MMFF-AM models are performed staged matching facilitated by STSC. This intelligent implementation allows SM-AM or MMFF-AM to tackle instances with different tracking difficulties, maintaining the operational efficiency and tracking accuracy of trade-offs. Our approach demonstrates the effectiveness of the proposed modules through comprehensive ablation experiments and the results obtained in the KITTI and nuScenes benchmark show that our method attains state-of-the-art performance in 3D MOT.

Optimized Deep LSTM-Based Multi-Object Tracking with Occlusion Handling Mechanism

The multi-object tracking is a basic computer vision process having a huge class of real-life tools that range from monitoring of medical video to surveillance. The goal of tracking numerous items is to place numerous objects in a scene, handle their identities throughout time, and construct trajectories for analysis. However, this is a complex task, because of certain issues like occlusions, complicated object dynamics, and variations in the appearance of objects. In this research, a new technique named TPRO-based Deep LSTM is developed for tracking multi-object with occlusion handling. Here, the videos are considered as input wherein the extraction of frames is done from each video. Each frame undergoes pre-processing with filtering to eliminate noise from frames. By using a sparse Fuzzy c-Means (FCM) and Local Optimal-Oriented Pattern (LOOP) features, the localization of objects is done. Moreover, the visual and spatial trackings are considered for hybrid tracking. The second derivative model and neighborhood search model are used to perform visual tracking. Then the occlusion handling is performed. Concurrently, with the use of Deep Long Short-Term Memory (Deep LSTM) the spatial tracking is performed and the Taylor Poor Rich Optimization (TPRO) algorithm assigns the weight and bias of the Deep LSTM. The TPRO is obtained by the unification of the Taylor series along with the Poor and Rich Optimization algorithm. By combining visual and spatial tracking, the final tracked output is generated. The devised method achieves a performance with the highest value of 88.9% for Multiple Object Tracking Precision (MOTP), smallest tracking distance (TD) of 4.185, average MOTP of 0.889, average TD of 4.201, and highest tracking number (TN) of 14.

Gesture-Controlled Robotic Arm for Agricultural Harvesting Using a Data Glove with Bending Sensor and OptiTrack Systems.

This paper presents a gesture-controlled robotic arm system designed for agricultural harvesting, utilizing a data glove equipped with bending sensors and OptiTrack systems. The system aims to address the challenges of labor-intensive fruit harvesting by providing a user-friendly and efficient solution. The data glove captures hand gestures and movements using bending sensors and reflective markers, while the OptiTrack system ensures high-precision spatial tracking. Machine learning algorithms, specifically a CNN+BiLSTM model, are employed to accurately recognize hand gestures and control the robotic arm. Experimental results demonstrate the system's high precision in replicating hand movements, with a Euclidean Distance of 0.0131 m and a Root Mean Square Error (RMSE) of 0.0095 m, in addition to robust gesture recognition accuracy, with an overall accuracy of 96.43%. This hybrid approach combines the adaptability and speed of semi-automated systems with the precision and usability of fully automated systems, offering a promising solution for sustainable and labor-efficient agricultural practices.

Indoor AR Navigation Framework Based on Geofencing and Image-Tracking with Accumulated Error Correction

This study presents a novel framework for improving indoor augmented reality (AR) navigation with modern smartphone technology, which is achieved by addressing two major challenges: managing large absolute coordinate spaces and reducing error accumulation in camera-based spatial tracking. Our contribution is significant in two ways. First, we integrate geofencing with indoor navigation by considering spatial tracking errors, timing for audio guidance, and dynamic 3D arrow visualization for effective local-to-global spatial coordinate transformation. This method achieves precise local positioning and seamlessly integrates with larger spatial contexts, overcoming the limitations of current AR systems. Second, we introduce a periodic image-based calibration approach to minimize the inherent error accumulation in camera-based tracking, enhancing accuracy over longer distances. Unlike prior studies focusing on individual technologies, our work explores the software architecture of indoor AR navigation by providing a comprehensive framework for its design and practical use. The practicality of our approach is validated through the implementation of a smartphone application at the Mineral Industry Museum of Akita University, highlighting the limitations of component technologies and demonstrating our framework’s effectiveness.

Development of Augmented Reality Platform Using Image Processing with Deep Learning Techniques

The development and structuring of an augmented reality (AR) platform integrating image processing, deep learning, and spatial tracking techniques have yielded promising results. AR technology, which enhances real-world objects with computer-generated information, holds vast potential for education, entertainment, and industry. Image processing enables real-time object detection, recognition, and tracking within the user's environment, while deep learning models enhance pattern recognition accuracy. Spatial tracking techniques ensure seamless integration of virtual content with the physical world, providing immersive AR experiences. Ant colony optimization (ACO) algorithms further enhance the AR platform's functionality by optimizing object placement and interaction. ACO mimics ant behavior to find optimal paths through graphs, effectively optimizing spatial layout and user interaction patterns in dynamic environments. The integration of ACO into the AR platform facilitates efficient pathfinding for augmented objects and enhances the system's responsiveness to user inputs. The successful implementation of the AR platform underscores its potential to revolutionize various industries, including healthcare, education, and marketing. By leveraging AI-driven algorithms and optimization techniques, AR systems can deliver personalized and engaging experiences that blur the boundaries between the digital and physical worlds. These results pave the way for future advancements in AR technology, driving innovation and reshaping human-computer interaction paradigms.

An omnidirectional spatial monocular visual localization and tracking method for indoor unmanned aerial vehicles based on the two-axis rotary table

Aiming at the complexity and poor adaptability of the calibration process in the traditional unmanned aerial vehicles (UAV) indoor visual positioning, this paper proposes an omnidirectional spatial tracking and localization method for indoor UAV based on the two-axis rotary table. Firstly, the position of the UAV fuselage feature points in the camera coordinate system of the turntable camera is computed by the Pespective-n-Point algorithm utilizing known position information of a plurality of feature points with pixel coordinate information in the corresponding image. Pixel coordinate information is extracted by obtaining UAV body-specific feature points from rotary table camera shots. Then, UAV localization in omnidirectional space can be obtained by using the calibrated rotary axis parameters of the rotary table and the rotation angle of the rotary table and substituting them into Rodriguez’s formula to unify the UAV position information acquired by the rotary table camera at different positions into a unified coordinate system. Finally, the angle at which the rotary table should rotate is calculated from the obtained UAV pose and the spatial position of the camera optical center and the rotary axis of the rotary table. The calculated angle is fed back to the turntable as feedback information. The rotary table receiving the feedback information is rotated to a position where the UAV is located at the center of the camera image. Thereby the tracking and localization of the UAV is realized. The experimental results show that the spatial range of localization is greatly expanded with the localization accuracy reaching the level of binocular visual localization. The omnidirectional spatial tracking and localization of indoor UAV can be conveniently realized by this method.

Abstract 4217: An engineered breast tumor microenvironment model, with single-cell spatial resolution, to assess spatial dynamics of tumor evolution

Abstract Cellular heterogeneity is a prominent feature of the tumor microenvironment, and the organization of these cells has been linked to clinical outcomes such as disease progression and drug resistance. While retrospective analysis of patient tumors has produced a myriad of insights, samples are limited and highly complex making it challenging to validate causality. Here we present an engineered breast tumor microenvironment model, with single cell spatial resolution, to systematically identify which cell phenotypes and their spatial arrangements may be driving ductal carcinoma in situ disease progression. A microfluidic dispenser (Biopixlar, Fluicell) was optimized to enable the spatial patterning of single cells to replicate native histology. To demonstrate the high spatial precision of our method, we first replicated an annotated 2D section of a breast tumor biopsy region of interest (ROI) using MCF10A, MDA-MB-231, primary mammary fibroblasts, THP-1 derived macrophages, and primary mesenchymal stem cells. The XY coordinate of each cell was identified to create a print map matching the original biopsy. Deposited cells adhered to their intended target with an average print fidelity of 2.4 µm. Next, the stromal compartment of the ROI was altered to include phenotypes associated with tumor promoting microenvironments (i.e. cancer associated fibroblasts, M2 macrophages) and the core of the MCF10A ring filled with MDA-MB-231. Each deposited cell contained a fluorescent tag corresponding with its cell type to facilitate real time spatial tracking and annotation. The tumor microenvironments were live cell imaged for 24 hours at 5 minute intervals. MCF10A cells spread to form their junctions and create a cohesive ring, MDA-MB-231 cells proliferated and rapidly moved around within the confined space, and the cells of the stromal compartment spread to create a dense tissue like structure. The microenvironments were cultured for an additional 2-6 days, fixed, and stained for markers of proliferation, phenotype shifts, cell-cell junctions, and morphology. Unlocking the relative contributions of the specific cell types in ductal carcinoma in situ tumor microenvironment to the overall evolution of invasive breast cancer may hold the answer for many questions in early detection, treatment, and cancer avatar models. Citation Format: Haylie R. Helms, Kody A. Oyama, Alexander E. Davies, Ellen M. Langer, Luiz E. Bertassoni. An engineered breast tumor microenvironment model, with single-cell spatial resolution, to assess spatial dynamics of tumor evolution [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 4217.

Unveiling the impact of machine learning algorithms on the quality of online geocoding services: a case study using COVID-19 data

AbstractIn today's era, the address plays a crucial role as one of the key components that enable mobility in daily life. Address data are used by global map platforms and location-based services to pinpoint a geographically referenced location. Geocoding provided by online platforms is useful in the spatial tracking of reported cases and controls in the spatial analysis of infectious illnesses such as COVID-19. The first and most critical phase in the geocoding process is address matching. However, due to typographical errors, variations in abbreviations used, and incomplete or malformed addresses, the matching can seldom be performed with 100% accuracy. The purpose of this research is to examine the capabilities of machine learning classifiers that can be used to measure the consistency of address matching results produced by online geocoding services and to identify the best performing classifier. The performance of the seven machine learning classifiers was compared using several text similarity measures, which assess the match scores between the input address data and the services' output. The data utilized in the testing came from four distinct online geocoding services applied to 925 addresses in Türkiye. The findings from this study revealed that the Random Forest machine learning classifier was the most accurate in the address matching procedure. While the results of this study hold true for similar datasets in Türkiye, additional research is required to determine whether they apply to data in other countries.

Temporal and spatial tracking of ultrafast light-induced strain and polarization modulation in a ferroelectric thin film.

Ultrashort light pulses induce rapid deformations of crystalline lattices. In ferroelectrics, lattice deformations couple directly to the polarization, which opens the perspective to modulate the electric polarization on an ultrafast time scale. Here, we report on the temporal and spatial tracking of strain and polar modulation in a single-domain BiFeO3 thin film by ultrashort light pulses. To map the light-induced deformation of the BiFeO3 unit cell, we perform time-resolved optical reflectivity and time-resolved x-ray diffraction. We show that an optical femtosecond laser pulse generates not only longitudinal but also shear strains. The longitudinal strain peaks at a large amplitude of 0.6%. The access of both the longitudinal and shear strains enables to quantitatively reconstruct the ultrafast deformation of the unit cell and to infer the corresponding reorientation of the ferroelectric polarization direction in space and time. Our findings open new perspectives for ultrafast manipulation of strain-coupled ferroic orders.

38 Application of Real-Time Location Systems for Automated Behavioral Monitoring in Sheep

Abstract Evaluating how, when, and where an animal moves in their environment can provide insights into their health and welfare states. Spatial tracking systems for monitoring sheep are less well developed than accelerometer-based systems, but have the potential to add important behavioral information to Precision Livestock Farming (PLF) applications. Global Positioning Systems (GPS) can be used to track pasture-based animals, but are not suitable to establish indoor locations. Real-Time Location Systems (RTLS) are a recent development that use radio-frequency signals emitted by tags to track the indoor location of an animal online and continuously. Raw RTLS data in the form of x, y coordinates can be used to calculate multiple spatial metrics, including velocity, acceleration, distance moved, the proportion of time an animal spends in different functional areas, social network attributes, and range size. RTLS can also be integrated with accelerometers to potentially improve predictive accuracy of behavioral algorithms. RTLS have been primarily tested and validated for indoor tracking of dairy cattle and have been used to assess behavioral time budgets and to detect space-use changes related to estrus or health status. For on-farm applications, these data must be transformed into threshold-based algorithms to generate notifications on changes in the health or welfare status of an animal. Algorithm development needs to account for confounding factors such as breed, sex, age, parity, and stage of lactation, as well as environmental variability that can influence the performance of decision-making tools under real-life conditions. In addition, independent validation studies of commercial systems are lacking, and many calculate positioning error with stationary objects rather than moving animals, potentially inflating accuracy estimates. In conclusion, there is growing empirical support for the use of RTLS as a PLF technology for on-farm animal health and welfare assessment. However, research is needed to validate these systems in sheep and identify spatial metrics that are accurate and robust predictors of health and welfare states for commercial applications.

Receiver architecture for on-off-keying free-space optical communications with channel estimation in the presence of residual spatial tracking error

Receiver architecture for on-off-keying free-space optical communications with channel estimation in the presence of residual spatial tracking error

Cortical temporal integration can account for limits of temporal perception: investigations in the binaural system

The auditory system has exquisite temporal coding in the periphery which is transformed into a rate-based code in central auditory structures, like auditory cortex. However, the cortex is still able to synchronize, albeit at lower modulation rates, to acoustic fluctuations. The perceptual significance of this cortical synchronization is unknown. We estimated physiological synchronization limits of cortex (in humans with electroencephalography) and brainstem neurons (in chinchillas) to dynamic binaural cues using a novel system-identification technique, along with parallel perceptual measurements. We find that cortex can synchronize to dynamic binaural cues up to approximately 10 Hz, which aligns well with our measured limits of perceiving dynamic spatial information and utilizing dynamic binaural cues for spatial unmasking, i.e. measures of binaural sluggishness. We also find that the tracking limit for frequency modulation (FM) is similar to the limit for spatial tracking, demonstrating that this sluggish tracking is a more general perceptual limit that can be accounted for by cortical temporal integration limits.

In Vivo Photoacoustic Monitoring of Stem Cell Location and Apoptosis with Caspase-3-Responsive Nanosensors.

Stem cell therapy has immense potential in a variety of regenerative medicine applications. However, clinical stem cell therapy is severely limited by challenges in assessing the location and functional status of implanted cells in vivo. Thus, there is a great need for longitudinal, noninvasive stem cell monitoring. Here we introduce a multidisciplinary approach combining nanosensor-augmented stem cell labeling with ultrasound guided photoacoustic (US/PA) imaging for the spatial tracking and functional assessment of transplanted stem cell fate. Specifically, our nanosensor incorporates a peptide sequence that is selectively cleaved by caspase-3, the primary effector enzyme in mammalian cell apoptosis; this cleavage event causes labeled cells to show enhanced optical absorption in the first near-infrared (NIR) window. Optimization of labeling protocols and spectral characterization of the nanosensor in vitro showed a 2.4-fold increase in PA signal from labeled cells during apoptosis while simultaneously permitting cell localization. We then successfully tracked the location and apoptotic status of mesenchymal stem cells in a mouse hindlimb ischemia model for 2 weeks in vivo, demonstrating a 4.8-fold increase in PA signal and spectral slope changes in the first NIR window under proapoptotic (ischemic) conditions. We conclude that our nanosensor allows longitudinal, noninvasive, and nonionizing monitoring of stem cell location and apoptosis, which is a significant improvement over current end-point monitoring methods such as biopsies and histological staining of excised tissue.

Modeling and Analyzing Xylem Vulnerability to Embolism as an Epidemic Process.

Xylem vulnerability to embolism can be quantified by "vulnerability curves" (VC) that are obtained by subjecting wood samples to increasingly negative water potential and monitoring the progressive loss of hydraulic conductivity. VC are typically sigmoidal, and various approaches are used to fit the experimentally obtained VC data for extracting benchmark data of vulnerability to embolism. In addition to such empirical methods, mechanistic approaches to calculate embolism propagation are epidemic modeling and network theory. Both describe the transmission of "objects" (in this case, the transmission of gas) between interconnected elements. In network theory, a population of interconnected elements is described by graphs in which objects are represented by vertices or nodes and connections between these objections as edges linking the vertices. A graph showing a population of interconnected xylem conduits represents an "individual" wood sample that allows spatial tracking of embolism propagation. In contrast, in epidemic modeling, the transmission dynamics for a population that is subdivided into infection-relevant groups is calculated by an equation system. For this, embolized conduits are considered to be "infected," and the "infection" is the transmission of gas from embolized conduits to their still water-filled neighbors. Both approaches allow for a mechanistic simulation of embolism propagation.

Gamified assessment of cognitive performance during moderate hypoxia.

There is a need for rapid and objective assessment tools to identify people at risk of impaired cognitive function during hypoxia. To test the hypotheses that performance on gamified cognitive tests examining the cognitive domains of executive function (Gridshot), working memory (Capacity) and spatial tracking (Multitracker) will be reduced during normobaric exposure to moderate normobaric hypoxia. Following three consecutive days of practice, twenty-one healthy adults (27 ± 5 y, 9 females) completed five 1-min rounds of the tablet-based games Gridshot, Capacity, and Multitracker (Statespace Labs, Inc.) at Baseline and 60 and 90 min after exposure to 14.0 ± 0.2% (hypoxia) and 20.6 ± 0.3% (normoxia) oxygen. Both conditions were completed on the same day and were administered in a single-blind, block randomized manner. Arterial oxyhemoglobin saturation was estimated via forehead pulse oximetry (SpO2). Data were analyzed using ANCOVA with a covariate of Baseline. Compared to normoxia (98 ± 1%), SpO2 was lower (p < 0.001) at 60 (91 ± 3%) and 90 (91 ± 2%) min of hypoxia. No condition x time interaction effects were identified for any gamified cognitive tests (p ≥ 0.32). A main effect of condition was identified for Capacity (p = 0.05) and Multitracker (p = 0.04), but not Gridshot (p = 0.33). Post hoc analyses of the composite scores for both Capacity (p = 0.11) and Multitracker (p = 0.73) demonstrated no difference between conditions. Performance on gamified cognitive tests was not consistently affected by acute normobaric moderate hypoxic exposure.

The design of a flexible RF generator for driving Acousto-Optical devices in space applications

Acousto-Optical (AO) devices are not only used for filtering purposes in spectral imaging, but also in optical communications and spatial tracking systems. Some AO devices with space applications are AO Tunable Filters (AOTFs), Modulators (AOMs), Deflectors (AODs) and Frequency Shifters (AOFSs). Though these device’s applications differ, they are all controlled with Radio-Frequency (RF) signals. These signals are converted by a transducer into an acoustic wave, which propagates inside the AO device. The interaction between the incoming light and the acoustic waves inside the birefringent AO material creates multiple output beams. This interaction can result in filtering, modulation, deflection or frequency shifting, depending on the AO device in question. This research focuses on the design of a flexible, uniform RF generator, applicable to all AO devices in the space applications domain. The RF output maximizes the performance of the AO device, while the use of components available in space qualified grades eases integration with future space missions. Its design is a key step towards a miniaturized, space qualified, general-purpose RF generator. This research presents schematics, design and preliminary component test results.

Nail it! vision-based drift correction for accurate mixed reality surgical guidance.

Mixed reality-guided surgery through head-mounted displays (HMDs) is gaining interest among surgeons. However, precise tracking of HMDs relative to the surgical environment is crucial for successful outcomes. Without fiducial markers, spatial tracking of the HMD suffers from millimeter- to centimeter-scale drift, resulting in misaligned visualization of registered overlays. Methods and workflows capable of automatically correcting for drift after patient registration are essential to assuring accurate execution of surgical plans. We present a mixed reality surgical navigation workflow that continuously corrects for drift after patient registration using only image-based methods. We demonstrate its feasibility and capabilities using the Microsoft HoloLens on glenoid pin placement in total shoulder arthroplasty. A phantom study was conducted involving five users with each user placing pins on six glenoids of different deformity, followed by a cadaver study by an attending surgeon. In both studies, all users were satisfied with the registration overlay before drilling the pin. Postoperative CT scans showed 1.5mm error in entry point deviation and 2.4[Formula: see text] error in pin orientation on average in the phantom study and 2.5mm and 1.5[Formula: see text] in the cadaver study. A trained user takes around 90s to complete the workflow. Our method also outperformed HoloLens native tracking in drift correction. Our findings suggest that image-based drift correction can provide mixed reality environments precisely aligned with patient anatomy, enabling pin placement with consistently high accuracy. These techniques constitute a next step toward purely image-based mixed reality surgical guidance, without requiring patient markers or external tracking hardware.

Four-dimensional reconstruction and characterization of bladder deformations

Background and ObjectivePelvic floor disorders are prevalent diseases and patient care remains difficult as the dynamics of the pelvic floor remains poorly understood. So far, only 2D dynamic observations of straining exercises at excretion are available in the clinics and 3D mechanical defects of pelvic organs are not well studied. In this context, we propose a complete methodology for the 3D representation of non-reversible bladder deformations during exercises, combined with a 3D representation of the location of the highest strain areas on the organ surface. MethodsNovel image segmentation and registration approaches have been combined with three geometrical configurations of up-to-date rapid dynamic multi-slice MRI acquisitions for the reconstruction of real-time dynamic bladder volumes. ResultsFor the first time, we proposed real-time 3D deformation fields of the bladder under strain from in-bore forced breathing exercises. The potential of our method was assessed on eight control subjects undergoing forced breathing exercises. We obtained average volume deviations of the reconstructed dynamic volume of bladders around 2.5% and high registration accuracy with mean distance values of 0.4 ± 0.3 mm and Hausdorff distance values of 2.2 ± 1.1 mm. ConclusionsThe proposed framework provides proper 3D+t spatial tracking of non-reversible bladder deformations. This has immediate applicability in clinical settings for a better understanding of pelvic organ prolapse pathophysiology. This work can be extended to patients with cavity filling or excretion problems to better characterize the severity of pelvic floor pathologies or to be used for preoperative surgical planning.

Timber–Prairie Wolf Optimization-Dependent Deep Learning Classifier for Anomaly Detection in Surveillance Videos

Anomaly detection in public places using the video surveillance gains significance due to the real-time monitoring and security that ensures the personal assets and public security. Accordingly, in this research, a deep CNN model with Timber–Prairie wolf optimization algorithm (TPWO) optimization is proposed for surveillance-based anomaly detection. To support the TPWO-based deep CNN anomaly detection model, tracking model named OptSpatio tracking model tracks the location and movement of the anomalous objects in any locality. The OptSpatio tracking model uses both visual and spatial tracking models to monitor any anomalous activity. On the other hand, TPWO is designed to tune the deep classifier for acquiring better detection performance. The TPWO-based model surpasses the competent methods in terms of accuracy by 97.214%, sensitivity by 97.831% and specificity by 96.668% with minimal EER of 2.786%. The MOTP values are also obtained at a rate of 0.7325; moreover, the effectiveness of the TPWO method is justified at the object-, frame-, and pixel-level analysis.