Rapid detection and precise segmentation of the weld seam region of interest (ROI) remain a core challenge in robotic intelligent grinding. To address this issue, this paper proposes a method for weld seam ROI detection and segmentation based on the fusion of active and passive vision. The proposed approach primarily consists of two stages: weld seam image instance segmentation and weld seam ROI point cloud segmentation. In the image segmentation stage, an enhanced segmentation network is constructed by integrating a convolutional attention module into YOLOv8n-seg, which effectively improves the localization accuracy and mask extraction quality of the weld seam region. In the point cloud segmentation stage, the 3D point cloud is first mapped onto a 2D pixel plane to achieve spatial alignment. Subsequently, a coarse screening of the projected point cloud is performed based on the bounding boxes output from the instance segmentation, eliminating a large amount of redundant data. Furthermore, a grayscale matrix is constructed based on the segmentation masks, enabling precise extraction of the weld seam ROI point cloud through point-wise discrimination. Experimental results demonstrate that the proposed method achieves high-quality segmentation of the weld seam region, providing a reliable foundation for robotic automated grinding.

Governance in the domain of Information Technology (IT) has gradually progressed into a strategic asset that empowers universities to synchronize and realign their technological aspirations with their institutional goals. In doing this, universities seek to maintain accountability in their decision-making processes. Within the Namibian higher education sector, the task of developing a functional IT governance framework entails more than merely applying technical tools. This is because it also requires strong leadership capabilities to ensure the relevance of such a framework. This research explored and examined how institutional leaderships facilitate IT governance within three prominent universities in Namibia. Using the mixed-methods approach, with data collected from the directors and professionals within IT departments of the universities and an amalgam of the Upper Echelon and Network Governance theories, the study explains how leadership traits, strategic institutional visions, and relational skills influence governance of IT in the universities. The findings of the study revealed that elements such as the dedicated leadership, digital proficiency, and collaboration of efforts and activities are crucial in developing an effective IT governance framework within the universities. The key conclusion of the study is that institutional leadership facilitates effective IT governance through elements such as shared vision, staff empowerment, and coordination of activities that synch structures of governance with the goals of universities. Based on this conclusion, the study recommends for upskilling of the institutional leaders, making IT governance a mandatory aspect within institutional structures, and installation of collaborative, yet, dynamic governance systems that promote accountability and innovation across institutional departments.

The role of the cerebellum has long been thought to be limited to sensorimotor processes. Recently, its involvement in a broader set of cognitive and associative tasks has challenged this view. Recent studies have expanded the cerebellum's functional repertoire into the visual domain, by identifying three topographically organized clusters exhibiting visual spatial responses using the population receptive field model. In this experiment, researchers used a simple retinotopic mapping stimulus during strict fixation. This represents a situation very different from our everyday vision, which is characterized by continual eye movements, and complex naturalistic visual stimulation. This makes it hard to translate the previous results to natural, active vision. Here, we used topographic connectivity from V1 to investigate the visual topographic organization in the human cerebellum (of either sex) and its dependence on cognitive state (comparing movie watching and resting state experiments). We find that movie watching evokes visual representations with a clear eccentricity gradient in OMV that was not found in a simple retinotopic mapping experiment. We furthermore discovered a novel topographically organized area in the cerebellum, again evoked specifically during movie watching and not in resting state. This latter area is located in the cerebellar Crus II area and falls within regions usually assigned to the cerebellar default mode network. Our results show that we can reveal task-dependent properties of the visual organization when using different cognitive states and how this can provide information about the processing of visual information, also in regions not previously considered to be visually responsive.

It is reasonable to argue that researching imagery experiences requires substantial use of conceptual modeling. Cognitive architectures have been used to explain cognitive phenomena like perception, action, and information gathering, and to model them in computational solutions. However, the question arises: Is there a lack of cognitive architectures or models that represent relational and classificatory knowledge of imagery experiences? This systematic review defines the concepts of cognitive architecture and cognitive model and examines how recent research relates the concepts to imagery experiences. A concept token research methodology is applied in search of keywords and key phrases that signify occurrences of targeted concepts. The methodology is viewed as a way to define a research area based on the concept of mental imagery and other related concepts that expand this area. The results demonstrate a significant and steady upward trend in publications from the research area in the last few years. The concepts of mental imagery and motor imagery emerged as the most regularly discussed, while others, such as imagery experiences, sensorimotor, mental model and active vision, were addressed rather rarely and thus represent new avenues for investigation.

SUMMARYIn active vision, the brain receives and encodes discontinuous streams of visual information gated by saccadic eye movements. Saccadic modulation of neural activity is hypothesized to evolve to support perception and memory; however, it remains unclear whether this mechanism exist in humans, and potential functional roles have not been determined. We used eye tracking and intracranial local field potentials recorded from invasively monitored epilepsy patients when they performed visual encoding tasks, and observed consistent evoked potentials following saccades (saccade-related evoked potentials, SREPs) across the cerebrum. These SREPs were not attributable to ocular muscle activity or retinal input. Their magnitudes were not explained by spatial proximity to eye muscles or saccade eccentricity, and their polarity bore no relationship to saccade direction. Instead, the phase of pre-saccadic oscillation aligned with saccade timing and dissociated SREPs with positive polarity from those with negative polarity. Spatiotemporal profiling revealed that SREPs emerged earliest and with the greatest magnitude in the temporal lobes. We developed a saccade-related neural dynamic (SRND) model that characterized pre-saccadic oscillatory activity and SREP at each electrode contact location using finite features. Random forest models trained with these features achieved 62.6% balanced accuracy for predicting next-day recognition (long-term memory). Using the Shapley value, a framework for explaining machine learning models, we identified an SREP profile, characterized by earlier latency and larger magnitude, which was associated with successful visual encoding. In contrast, predicting saccade direction using the same SRND model performed at chance level, indicating that the observed SREP is less likely a corollary discharge signaling saccadic motor copy. These findings demonstrate SREPs as a neural mechanism of saccadic modulation with a role in human visual encoding.

The appreciation of parallel visual pathways from retina to cortex in mammalian vision has greatly advanced our understanding of sensory processing. While these anatomical pathways are well-characterized, their specific functional roles remain incompletely understood, particularly in terms of parsing natural scenes. Here, we start with a re-evaluation of the functional contributions of parallel X and Y pathways in the cat, emphasizing the underappreciated role of Y cells in spatial vision. Contrary to the traditional view that prioritizes the importance of high acuity for visual function and thus the role of X cells, because they support high acuity vision, we argue that lower spatial frequencies are more important for visual function. For instance, these lower frequencies dominate natural visual scenes, and Y cells are much more responsive to these than X cells are. We provide further anatomical and behavioral evidence that the Y pathway plays the dominant role in basic spatial vision, particularly during active vision. These findings challenge the adequacy of visual acuity and the X pathway as a proxy for visual function and underscore the need to consider functioning in response to natural scenes in models of visual processing. Implications for primate and rodent vision are discussed, including the need for comparative and integrative research and the importance of understanding similarities and differences across species.

Psychedelic compounds have the ability to generate altered states of consciousness and profoundly distort perception, often resulting in visual hallucinations. While psychedelics have recently regained attention for their potential cognitive and therapeutic effects, how these drugs affect visual processing to generate visual distortions and hallucinations is not as well characterized. Furthermore, studies investigating the effect of psychedelics on visual function have all been performed using head-fixed preparations, preventing animals from engaging in the natural visual behaviors that the visual system evolved to support. To determine the impact of psychedelics on active vision, we recorded neural activity in primary visual cortex (V1) of mice during free movement, while simultaneously recording eye and head position, before and after administration of the serotonergic psychedelic DOI (2,5-dimethoxy-4-iodoamphetamine). We find that DOI increases the frequency of visual active sensing behaviors during free movement and leads to a net reduction in the visually-evoked activity that these behaviors generate in V1. The effect of DOI was highly diverse across the population of V1 neurons, driving suppression and facilitation of visual responses in a laminar specific manner. Finally, the effects of DOI were dependent upon both visual input and the statistics of the stimulus. We found a striking dissociation between impact on gaze shift responses and flashed sparse noise presented during head-fixation, which may reflect predictability of the stimulus. These findings provide insights into how psychedelics disrupt sensory processing and the neural mechanisms underlying these altered perceptual states.

Neural dynamics in superior colliculus of freely moving mice.

Real-time identification and measurement algorithm of typical welding defects based on line-structured light active vision

Visual exploration-where the eyes move and when-is guided by prior experiences. Memory-guided viewing behavior is altered in healthy aging and is further disrupted in amnestic mild cognitive impairment (aMCI), a condition in which there is accelerated structural and functional decline of the hippocampus and associated medial temporal lobe structures (HC/MTL). Computational modeling has demonstrated the potential for rapid information flow from the HC/MTL to regions responsible for the cognitive control of eye movements, such that visual exploration behavior could be impacted in the moment and on an ongoing fashion. It was predicted here, then, that older adults and individuals with aMCI would show changes in naturalistic viewing compared to younger adults, even in the absence of any memory task. Multivariate analyses revealed that viewing for younger adults was characterized by larger saccade amplitudes and a larger area of exploration; the opposite pattern was reliably expressed by individuals with aMCI. Viewing patterns of healthy older adults were associated with shorter gaze durations. The entropy of viewing in older adults was associated with overall cognitive status, as determined by the Montreal Cognitive Assessment, highlighting the top-down influence of cognitive function on active vision. Lower scores on the memory subtest were reliably associated with a pattern of viewing characterized by fewer fixations (with longer durations), saccades, regions explored, smaller area of exploration, and lower entropy, mimicking some of the viewing features of the aMCI group and suggesting that increasing HC/MTL decline results in less exploratory viewing patterns. These findings reveal the ongoing influence of the hippocampus and its extended system on moment-to-moment naturalistic viewing.

Introduction: Active vision therapy, integrating perceptual learning with dichoptic or binocular environments, has shown potential effectiveness in treating amblyopia. However, uncertainties remain regarding the optimal types of stimuli and the best approaches and sequences for their delivery. This systematic review aimed to evaluate the effectiveness of psychophysical visual-stimuli-based interventions, particularly perceptual learning and dichoptic training, in treating amblyopia. Materials and Methods: A comprehensive literature search across major databases, such as PubMed and Google Scholar, yielded 26 studies involving 993 patients with amblyopia. These studies investigated various visual training methods, including perceptual learning, dichoptic stimulation, and combinations of both, using stimuli, such as Gabor patches, letter optotypes, Vernier stimuli, and random-dot stereograms. Results: The findings indicate that perceptual learning enhances visual acuity, contrast sensitivity, and stereopsis by leveraging neural plasticity, even in adult patients. Dichoptic training, which engages both eyes simultaneously, shows promise in reducing suppression and improving binocular integration, offering potential advantages over traditional patching therapy. Gabor patches emerged as particularly effective, stimulating the visual cortex to drive neural efficiency. Conclusion: Vision therapy is an effective strategy for treating amblyopia and may reduce overall treatment time when used in conjunction with patching. In addition, it is crucial to tailor stimuli to match the individual characteristics of each patient during monocular and binocular training.

This article deals with the neural circuits involved in the generation and suppression of saccadic eye movements. Voluntary eye movements are known to be organized in two-dimensional horizontal and vertical coordinate system, whereas the vestibulo-ocular reflex (VOR) uses a three-dimensional semicircular canal coordinate system. However, it is generally accepted that the neural circuits for saccades and the quick phase of vestibular nystagmus share a common pathway. Despite the discrepancy, this issue has not been noticed, because the output pathway for the vertical saccade system remains poorly understood. To resolve this discrepancy, we analyzed the pathway from the superior colliculus (SC) to vertical ocular motoneurons, using intracellular recording and staining techniques. We found that the saccade system and the vestibuloocular system share the common semicircular coordinate system. Based on this saccade generation circuit, we proceeded to analyze the saccade trigger system. Our results showed that the saccade-triggering signal originates from the saccade-driving pathway in the SC, passing through inhibitory burst neurons, to suppress the tonic firing of inhibitory omnipause neurons in the raphe nucleus. This triggering mechanism is then discussed in relation to the suppression pathway for saccades during fixation and active vision.

Neural coding and circuitry of active vision in mice

The role of active vision in the primary visual cortex of freely-moving marmosets

Visual search is a ubiquitous task; people search for objects on a daily basis. However, the majority of the existing visual search literature focuses on passive search on a 2D computer screen, a far cry from emulating a real-world environment. Search is a real-world task that involves active observation. Search targets may be occluded, completely out of the observer's line of sight, or oriented in unconventional ways. This is typically mitigated by actively selecting viewpoints, an important aspect of search behaviour with limited scope on a computer screen. Our goal was to explore viewpoint selection in active visual search. Subject eye and head movements were tracked as they moved freely while searching for toy objects in a controlled 3-dimensional environment, yielding the first such record of search-driven viewpoint selection. We found that subjects utilized their full range of eye and head motion to move from viewpoint to viewpoint, apparently employing a variety of objectives including changing viewing height and pose depending on object 3D pose. Subjects were also adept at selecting unobstructed views to search through otherwise occluded areas with objects. Furthermore, subjects completed the search task with high accuracy, even with no training on the environment. Although no learning was found in terms of accuracy over the duration of the experiment, increases in efficiency were found for other metrics such as response time, number of fixations, and distance travelled, particularly in target present trials where the target was not visible from the starting location. These results paint the story of a visual system that selects and moves to useful and informative views to facilitate the successful execution of an active visual search task, and stresses the significance of active vision research in understanding how vision is used in naturalistic environments.

ABSTRACTActive object detection (AOD) enables a system to actively adjust camera parameters or plan the next viewpoint to improve detection accuracy when the current visual input is insufficient. However, most existing AOD methods assume that the target object is visible from the initial viewpoint, which is often unrealistic and reduces task efficiency. To address this limitation, we propose a novel AOD framework that leverages partial prior information to enhance detection performance and task efficiency. Specifically, we construct an extensible prior information library that describes large and easily identifiable adjacent objects (Adj‐objects) that are spatially related to the target. This allows the system to initiate AOD based on the presence of an Adj‐object, even when the target is initially out of view. Our approach incorporates a duelling deep Q‐learning network (Duelling‐DQN) with a newly designed reward function to effectively utilise prior information. Additionally, we introduce a viewpoint storage scheme to support fast retrieval and transition between viewpoints. We evaluate the proposed method on the Active Vision Dataset (AVD) and compare it with several state‐of‐the‐art (SOTA) approaches. The experimental results show that our method achieves a superior average success rate of 81.3%, demonstrating its effectiveness in overcoming the initial state limitations of traditional AOD tasks.

Bees' remarkable visual learning abilities make them ideal for studying active information acquisition and representation. Here, we develop a biologically inspired model to examine how flight behaviours during visual scanning shape neural representation in the insect brain, exploring the interplay between scanning behaviour, neural connectivity, and visual encoding efficiency. Incorporating non-associative learning-adaptive changes without reinforcement-and exposing the model to sequential natural images during scanning, we obtain results that closely match neurobiological observations. Active scanning and non-associative learning dynamically shape neural activity, optimising information flow and representation. Lobula neurons, crucial for visual integration, self-organise into orientation-selective cells with sparse, decorrelated responses to orthogonal bar movements. They encode a range of orientations, biased by input speed and contrast, suggesting co-evolution with scanning behaviour to enhance visual representation and support efficient coding. To assess the significance of this spatiotemporal coding, we extend the model with circuitry analogous to the mushroom body, a region linked to associative learning. The model demonstrates robust performance in pattern recognition, implying a similar encoding mechanism in insects. Integrating behavioural, neurobiological, and computational insights, this study highlights how spatiotemporal coding in the lobula efficiently compresses visual features, offering broader insights into active vision strategies and bio-inspired automation.

Optimized Design and Calibration of a Human-Eye-Sized Active Binocular Vision System Based on Spherical Parallel Mechanism

Barrier Lyapunov function based guidance law design for gap traversal of aerial vehicles.

Multiple object tracking with multi-view active vision to effectively find plant nodes in a cluttered tomato greenhouse