Spatial Information Processing Research Articles

Three-Dimensional CKANs: UUV Noncooperative Target State Estimation Approach Based on 3D Convolutional Kolmogorov–Arnold Networks

Accurate and stable estimation of the position and trajectory of noncooperative targets is crucial for the safe navigation and operation of sonar-equipped underwater unmanned vehicles (UUVs). However, the uncertainty associated with sonar observations and the unpredictability of noncooperative target movements often undermine the stability of traditional Bayesian methods. This paper presents an innovative approach for noncooperative target state estimation utilizing 3D Convolutional Kolmogorov–Arnold Networks (3DCKANs). By establishing a non-Markovian model that characterizes state estimation of UUV noncooperative targets under uncertain observations, we leverage historical data to construct 3D Convolutional Kolmogorov–Arnold Networks. This network learns the patterns of sonar observations and target state transitions from a substantial offline dataset, allowing it to approximate the posterior probability distribution derived from past observations effectively. Additionally, a sliding window technique is integrated into the convolutional neural network to enhance the estimator’s fault tolerance with respect to observation data in both temporal and spatial dimensions, particularly when posterior probabilities are unknown. The incorporation of the Kolmogorov–Arnold representation within the convolutional layers enhances the network’s capacity for nonlinear expression and adaptability in processing spatial information. Finally, we present statistical experiments and simulation cases to validate the accuracy and stability of the proposed method.

Rockfall Mapping and Monitoring Across the Kalymnos Sport Rock Climbing Sites, Based on Ultra-High-Resolution Remote Sensing Data and Integrated Simulations

This manuscript presents a multidisciplinary study that proposes a methodology for delineating and categorizing vulnerability at rockfall risk areas to avoid human injuries and infrastructure damage caused by rockfalls. The presented workflow includes (i) classical geological mapping, (ii) the interpretation of high-resolution satellite data for observing the spatial distribution of fallen boulders, (iii) analytical hierarchy processing of spatial information within a Geographical Information System (GIS) platform, (iv) close-range remote sensing campaigns with Unmanned Aerial Systems (UASs), and (v) integrated simulation of rockfall events. This methodology was applied to Kalymnos Island, which belongs to the Dodecanese Islands complex of the southeastern Aegean Sea in Greece. It is characterized by unique geomorphological features, including extensive vertical limestone cliffs that span the island. These cliffs make it one of the world’s most densely concentrated areas for sport climbing. The results highlighted the areas that the local authorities need to focus on and suggested measures for increasing the safety of climbers and infrastructure.

Projection-specific circuits of retrosplenial cortex with differential contributions to spatial cognition.

Retrosplenial cortex (RSC) is a brain region involved in neuropsychiatric and neurodegenerative disorders. It has reciprocal connections with a diverse set of cortical and subcortical brain regions, but the afferent structure and behavioral function of circuits defined by its projection-specific sub-populations have yet to be determined. The corticocortical connections between RSC and secondary motor cortex (M2), as well as corticothalamic connections between RSC and anterodorsal thalamus (AD) have been hypothesized to function as semi-independent, but parallel pathways that impact spatial information processing in distinct ways. We used retrograde and anterograde viral tracers and monosynaptic retrograde rabies virus to quantitatively characterize and compare the afferent and efferent distributions of retrosplenial neuron sub-populations projecting to M2 and AD. AD-projecting and M2-projecting RSC neurons overlap in their collateral projections to other brain regions, but not in their projections to M2 and AD, respectively. Compared with AD-projecting RSC neurons, M2-projecting RSC neurons received much greater afferent input from the dorsal subiculum, AD, lateral dorsal and lateral posterior thalamus, and somatosensory cortex. AD-projecting RSC neurons received greater input from the anterior cingulate cortex and medial septum. We performed chemogenetic inhibition of M2- and AD-projecting RSC neurons and examined its impact on object-location memory, object-recognition, open-field exploration, and place-action association. Our findings indicate that inhibition of M2-projecting RSC neurons impairs object location memory as well as place-action association, while the RSC to AD pathway impacts only object-location memory. The findings indicate that RSC is composed of semi-independent circuits distinguishable by their afferent/efferent distributions and differing in the cognitive functions to which they contribute.

Information transfer from spatial to social distance in rats: implications for the role of the posterior parietal cortex in spatial-social integration.

Humans and other social animals can represent and navigate complex networks of social relationships in ways that are suggestive of representation and navigation in space. There is some evidence that cortical regions initially required for processing space have been adapted to include processing of social information. One candidate region for supporting both spatial and social information processing is the posterior parietal cortex (PPC). We examined the hypothesis that rats can transfer or generalize distance information across spatial and social domains and that this phenomenon requires the PPC. In a novel apparatus, rats learned to discriminate two conspecifics positioned at different spatial distances (near vs. far) in a goal-driven paradigm. Following spatial learning, subjects were tested on probe trials in which spatial distance was replaced with social distance (cagemate vs. less familiar conspecific). The PPC was chemogenetically inactivated during a subset of probe sessions. We predicted that, in control probe trials, subjects would select conspecifics whose social distance matched the previously learned spatial distance. That is, if trained on the near distance, the rat would choose the highly familiar cagemate, and if trained on the far distance, the rat would choose the less familiar conspecific. Subjects learned to discriminate conspecifics based on spatial distance in our goal-driven paradigm. Moreover, choice for the appropriate social distance in the first probe session was significantly higher than chance. This result suggests that rats transferred learned spatial information to social contexts. Contrary to our predictions, PPC inactivation did not impair spatial to social information transfer. Possible reasons are discussed. To our knowledge, this is the first study to provide evidence that spatial and social distance are processed by shared cognitive mechanisms in the rat model.

Efficient Optimized YOLOv8 Model with Extended Vision.

In the field of object detection, enhancing algorithm performance in complex scenarios represents a fundamental technological challenge. To address this issue, this paper presents an efficient optimized YOLOv8 model with extended vision (YOLO-EV), which optimizes the performance of the YOLOv8 model through a series of innovative improvement measures and strategies. First, we propose a multi-branch group-enhanced fusion attention (MGEFA) module and integrate it into YOLO-EV, which significantly boosts the model's feature extraction capabilities. Second, we enhance the existing spatial pyramid pooling fast (SPPF) layer by integrating large scale kernel attention (LSKA), improving the model's efficiency in processing spatial information. Additionally, we replace the traditional IOU loss function with the Wise-IOU loss function, thereby enhancing localization accuracy across various target sizes. We also introduce a P6 layer to augment the model's detection capabilities for multi-scale targets. Through network structure optimization, we achieve higher computational efficiency, ensuring that YOLO-EV consumes fewer computational resources than YOLOv8s. In the validation section, preliminary tests on the VOC12 dataset demonstrate YOLO-EV's effectiveness in standard object detection tasks. Moreover, YOLO-EV has been applied to the CottonWeedDet12 and CropWeed datasets, which are characterized by complex scenes, diverse weed morphologies, significant occlusions, and numerous small targets. Experimental results indicate that YOLO-EV exhibits superior detection accuracy in these complex agricultural environments compared to the original YOLOv8s and other state-of-the-art models, effectively identifying and locating various types of weeds, thus demonstrating its significant practical application potential.

Coastline target detection based on UAV hyperspectral remote sensing images

Timely and accurate monitoring of typical coastal targets using remote sensing technology is crucial for maintaining marine ecological stability. Hyperspectral target detection technology proves to be an effective tool in extracting various typical materials along the coastline. Traditional target detection methods using spectral domain information can effectively retain the intrinsic properties of the material. However, it is difficult to effectively recognize targets in homogeneous regions by using only spectral domain information, which may lead to insufficient utilization of spatial information. In this study, a detector based on signal-to-noise ratio fusion constrained energy minimization with low-rank sparse decomposition (SFLRSD) is proposed. This detector improves the separability of background and target by obtaining spatial domain information from hyperspectral images and fusing spectral domain information. First, total variation regularization and fractional Fourier transform are applied to process spatial and spectral domain information, respectively. The constrained energy minimization (CEM) detector is used to improve the separability between the target and background of the processed data. Then, the background and anomalies are represented as low-rank and sparse components, respectively, using low-rank sparse matrix factorization. This transforms the model solution into a covariance matrix problem, which is then solved using marginal distance difference (MDD) to isolate anomalous parts. Subsequently, the anomaly parts are fused with CEM detector results, weighted by their respective signal-to-noise ratios. This detection model leverages unified hyperspectral image features, enhancing spectral discreteness of anomalous targets and backgrounds. Finally, experiments on custom created hyperspectral dataset show that the proposed method outperforms other baseline methods in terms of visualization and quantitative performance. In this paper, we not only propose a new hyperspectral target detection method, but we also collect three typical marine litter of different materials by means of airborne hyperspectral remote sensing and construct four hyperspectral datasets in a real environment. All the simulation experiments in this paper are conducted in these four datasets.

PROBLEMS OF MANAGING TECHNOLOGICAL REGIMES IN GAS DISTRIBUTION NETWORKS AND MODERN APPROACHES TO THEIR SOLUTION

This article systematically discusses the problems of managing technological regimes in gas distribution networks and modern approaches to their resolution. The scientific-theoretical and practical aspects of managing gas distribution networks (GDN) are analyzed. Based on the obtained results, a scientifically grounded comparative explanation of the existing contradictions in modern and traditional gas distribution networks is provided, and the role of key factors in the emergence of these contradictions is investigated. It is shown that technological backwardness in traditional GDNs increases energy consumption and operational costs, thereby reducing the economic efficiency of the system and increasing the risk of accidents. The research suggests decision-making methods for solving management problems in traditional GDNs by maintaining the technical-structural level and implementing telemetry systems. It is indicated that the widespread use of telemetry and telemechanics tools in the gas distribution industry enhances the safe delivery of gas to end-users and significantly reduces service costs. Another focus of the research is on the issues related to the processing of spatial information for the management of the technical components of gas distribution systems. The article also analyzes scientific research dedicated to the application of modern IT methods and tools in the management of GDNs, investigating management and forecasting models using neural networks and intelligent methods. Keywords: Gas distribution networks, modern GDN networks, traditional GDN networks, GIS systems, digital technologies, technogenic risks, neural networks, forecasting methods

Subpixel water area mapping based on spatial information processing at superpixel scale for multispectral imagery

ABSTRACT Subpixel water area mapping based on spatial information processing at superpixel scale (SISS) for multispectral imagery is proposed in this paper. In SISS, the superpixels with irregular shape are first obtained by segmenting the first principal component of the improved imagery which is improved from the the original coarse multispectral imagery by using bicubic interpolation; The random walk algorithm is then introduced to calculate these superpixels to obtain the spatial information. Finally, according to this spatial information, the label of the water area or the non-water area is assigned to each subpixel by the class allocation method, producing the final mapping result. The experimental results on two datasets show that the proposed method can achieve the best performance in both visual effects and quantitative indicators.

Behavioral Alterations of Spatial Cognition and Role of the Apolipoprotein E-ε4 in Patients with MCI Due to Alzheimer's Disease: Results from the BDSC-MCI Project.

Background: Beyond memory deterioration, spatial disorientation may occur along the continuum of normal aging-dementia of Alzheimer's type. The present study aims at detecting behavioral disorders of spatial cognition in prodromal Alzheimer's disease (AD) and verifying the association between Apolipoprotein E-ε4 (ApoE-ε4) genotype and gait patterns during a real-world naturalistic task. Methods: A sample of 58 elderly participants, of which 20 patients with mild cognitive impairment with CFS biomarker evidence of AD, 23 individuals with subjective cognitive decline (SCD), and 15 healthy controls (HCs), was tested by a modified version of the Detour Navigation Test (DNT-mv). Generalized linear models were run to explore the association between group belonging and wrong turns (WTs)/moments of hesitation (MsH) as behavioral disorientation scores of the DNT-mv as well as the effect of ApoE-ε4 genotype on time and walking speed registered by a smartphone app providing GPS tracking of body movement around urban environments. Results: Patients with MCI due to AD reported more WTs than individuals with SCD and HCs. Further, the ApoE-ε4 genotype determined a lower capacity in spatial information processing, influencing gait during naturalistic spatial navigation tasks. Conclusions: Behavior alterations of spatial cognition can be detected ecologically in prodromal AD. The use of technological solutions supporting gait analysis may help in corroborating the experimental observation.

Sunspot Group Detection and Classification by Dual Stream Convolutional Neural Network Method

The automatic detection and analysis of sunspots play a crucial role in understanding solar dynamics and predicting space weather events. This paper proposes a novel method for sunspot group detection and classification called the dual stream Convolutional Neural Network with Attention Mechanism (DSCNN-AM). The network consists of two parallel streams each processing different input data allowing for joint processing of spatial and temporal information while classifying sunspots. It takes in the white light images as well as the corresponding magnetic images that reveal both the optical and magnetic features of sunspots. The extracted features are then fused and processed by fully connected layers to perform detection and classification. The attention mechanism is further integrated to address the “edge dimming” problem which improves the model’s ability to handle sunspots near the edge of the solar disk. The network is trained and tested on the SOLAR-STORM1 data set. The results demonstrate that the DSCNN-AM achieves superior performance compared to existing methods, with a total accuracy exceeding 90%.

Intrinsic Bipolar Head-Direction Cells in the Medial Entorhinal Cortex.

Head-direction (HD) cells are a fundamental component in the hippocampal-entorhinal circuit for spatial navigation and help maintain an internal sense of direction to anchor the orientation in space. A classical HD cell robustly increases its firing rate when the head is oriented toward a specific direction, with each cell tuned to only one direction. Although unidirectional HD cells are reported broadly across multiple brain regions, computation modelling has predicted the existence of multiple equilibrium states of HD network, which has yet to be proven. In this study, a novel HD variant of bipolar HD cells in the medial entorhinal cortex (MEC) are identified that exhibit stable double-peaked directional tuning properties. The bipolar patterns remain stable in the darkness and across environments of distinct geometric shapes. Moreover, bipolar HD cells co-rotate coherently with unipolar HD cells to anchor the external visual cue. The discovery reveals a new spatial cell type of bipolar HD cells, whose unique activity patterns may comprise a potential building block for a sophisticated local neural circuit configuration for the internal representation of direction. These findings may contribute to the understanding of how the brain processes spatial information by shedding light on the role of bipolar HD cells in this process.

Hybrid attentive prototypical network for few-shot action recognition

Most previous few-shot action recognition works tend to process video temporal and spatial features separately, resulting in insufficient extraction of comprehensive features. In this paper, a novel hybrid attentive prototypical network (HAPN) framework for few-shot action recognition is proposed. Distinguished by its joint processing of temporal and spatial information, the HAPN framework strategically manipulates these dimensions from feature extraction to the attention module, consequently enhancing its ability to perform action recognition tasks. Our framework utilizes the R(2+1)D backbone network, coupling the extraction of integrated temporal and spatial features to ensure a comprehensive understanding of video content. Additionally, our framework introduces the novel Residual Tri-dimensional Attention (ResTriDA) mechanism, specifically designed to augment feature information across the temporal, spatial, and channel dimensions. ResTriDA dynamically enhances crucial aspects of video features by amplifying significant channel-wise features for action distinction, accentuating spatial details vital for capturing the essence of actions within frames, and emphasizing temporal dynamics to capture movement over time. We further propose a prototypical attentive matching module (PAM) built on the concept of metric learning to resolve the overfitting issue common in few-shot tasks. We evaluate our HAPN framework on three classical few-shot action recognition datasets: Kinetics-100, UCF101, and HMDB51. The results indicate that our framework significantly outperformed state-of-the-art methods. Notably, the 1-shot task, demonstrated an increase of 9.8% in accuracy on UCF101 and improvements of 3.9% on HMDB51 and 12.4% on Kinetics-100. These gains confirm the robustness and effectiveness of our approach in leveraging limited data for precise action recognition.

Differential functions of the dorsal and intermediate regions of the hippocampus for optimal goal-directed navigation in VR space

Goal-directed navigation requires the hippocampus to process spatial information in a value-dependent manner, but its underlying mechanism needs to be better understood. Here, we investigated whether the dorsal (dHP) and intermediate (iHP) regions of the hippocampus differentially function in processing place and its associated value information. Rats were trained in a place-preference task involving reward zones with different values in a visually rich virtual reality environment where two-dimensional navigation was possible. Rats learned to use distal visual scenes effectively to navigate to the reward zone associated with a higher reward. Inactivation of both dHP and iHP with muscimol altered the efficiency and precision of wayfinding behavior, but iHP inactivation induced more severe damage, including impaired place preference. Our findings suggest that the iHP is more critical for value-dependent navigation toward higher-value goal locations.

Differential functions of the dorsal and intermediate regions of the hippocampus for optimal goal-directed navigation in VR space.

Goal-directed navigation requires the hippocampus to process spatial information in a value-dependent manner, but its underlying mechanism needs to be better understood. Here, we investigated whether the dorsal (dHP) and intermediate (iHP) regions of the hippocampus differentially function in processing place and its associated value information. Rats were trained in a place-preference task involving reward zones with different values in a visually rich virtual reality environment where two-dimensional navigation was possible. Rats learned to use distal visual scenes effectively to navigate to the reward zone associated with a higher reward. Inactivation of both dHP and iHP with muscimol altered the efficiency and precision of wayfinding behavior, but iHP inactivation induced more severe damage, including impaired place preference. Our findings suggest that the iHP is more critical for value-dependent navigation toward higher-value goal locations.

Biological Basis and Computer Vision Applications of Image Phase Congruency: A Comprehensive Survey.

The concept of Image Phase Congruency (IPC) is deeply rooted in the way the human visual system interprets and processes spatial frequency information. It plays an important role in visual perception, influencing our capacity to identify objects, recognize textures, and decipher spatial relationships in our environments. IPC is robust to changes in lighting, contrast, and other variables that might modify the amplitude of light waves yet leave their relative phase unchanged. This characteristic is vital for perceptual tasks as it ensures the consistent detection of features regardless of fluctuations in illumination or other environmental factors. It can also impact cognitive and emotional responses; cohesive phase information across elements fosters a perception of unity or harmony, while inconsistencies can engender a sense of discord or tension. In this survey, we begin by examining the evidence from biological vision studies suggesting that IPC is employed by the human perceptual system. We proceed to outline the typical mathematical representation and different computational approaches to IPC. We then summarize the extensive applications of IPC in computer vision, including denoise, image quality assessment, feature detection and description, image segmentation, image registration, image fusion, and object detection, among other uses, and illustrate its advantages with a number of examples. Finally, we discuss the current challenges associated with the practical applications of IPC and potential avenues for enhancement.

Interaction between visual working memory and upright postural control in young adults: an event-related potential study based on the n-back paradigm.

As a part of the overall information-processing system of the brain, postural control is related to the cognitive processes of working memory. Previous studies have suggested that cognitive tasks and postural control processes can compete for resources in common brain areas, although there is an "inverted U" relationship between arousal level and behavioral control - the arousal level of individuals changes when performing cognitive tasks. However, the exact neural connections between the two are unclear. This may be related to the nature of cognitive tasks. Some studies believe that posture occupies not only spatial information processing resources but also visual non-spatial information processing resources. Other studies believe that posture control only occupies spatial information processing resources in the central system, but does not occupy non-spatial information processing resources. Previous studies used different cognitive task materials and reached different conclusions. In this study, we used the same visuospatial and non-spatial materials, the n-back visual working memory paradigm, the event-related potential technique to investigate the effects of visuospatial and non-spatial working memory tasks on adolescents' postural control under different cognitive loads. The results of this study showed that in both visuospatial and non-spatial conditions, the N1 effect of the parieto-occipital lobe was larger during upright posture than in the sitting position (160-180 ms), the P300 effect of the central parieto-occipital region (280-460 ms) was induced by working memory in different postures, and the P300 wave amplitude was higher in the sitting position than in the upright position. We demonstrated that upright postural control enhances early selective attention but interferes with central memory encoding, thus confirming that postural control and visuospatial and non-spatial working memory share brain regions and compete with each other.

Deciphering the Role of the Nucleus Accumbens Shell Area on Spatial Memory Deficits Induced by Neuropathic Pain in Rats

The nucleus accumbens shell (NAcSh) is a major structure associated with distinct aspects of reward and mnemonic information encoding, relying on spatial data to define optimal behavioral strategies. Chronic pain-derived striatal plasticity is considered one underpinning cause of working memory (WM) impairments. However, it is unclear how the NAcSh is involved in these spatial deficits. To address this, we evaluated the impact of unilateral local NAcSh electrical lesions during the execution of a food-reinforced eight-shaped spatial alternation WM task. Behavioral performance was assessed in rats after the onset of the neuropathic pain model—spared nerve injury (SNI). Our findings indicate that the induction of SNI and/or NAcSh lesions did not significantly impact the animals’ performance accuracy or motor activity during the execution of the behavioral task, but altered their response latency patterns. In addition, these manipulations did not induce significant antinociceptive effects. Collectively, these results suggest that the NAcSh may participate in specific aspects of spatial information integration and processing under neuropathic pain conditions.

EEG markers of successful allocentric spatial working memory maintenance in humans.

Several brain regions in the frontal, occipital and medial temporal lobes are known to contribute to spatial information processing. In contrast, the oscillatory patterns contributing to allocentric spatial working memory maintenance are poorly understood, especially in humans. Here, we tested twenty-three 21- to 32-year-old and twenty-two 64- to 76-year-old healthy right-handed adults in a real-world, spatial working memory task and recorded electroencephalographic (EEG) activity during the maintenance period. We established criteria for designating recall trials as perfect (no errors) or failed (errors and random search) and identified 8 young and 13 older adults who had at least 1 perfect and 1 failed trial amongst 10 recall trials. Individual alpha frequency-based analyses were used to identify oscillatory patterns during the maintenance period of perfect and failed trials. Spectral scalp topographies showed that individual theta frequency band relative power was stronger in perfect than in failed trials in the frontal midline and posterior regions. Similarly, gamma band (30-40 Hz) relative power was stronger in perfect than in failed trials over the right motor cortex. Exact low-resolution brain electromagnetic tomography in the frequency domain identified greater theta power in perfect than in failed trials in the secondary visual area (BA19) and greater gamma power in perfect than in failed trials in the right supplementary motor area. The findings of this exploratory study suggest that theta oscillations in the occipital lobe and gamma oscillations in the secondary motor cortex (BA6) play a particular role in successful allocentric spatial working memory maintenance.

Hybrid CNN-LSTM model driven image segmentation and roughness prediction for tool condition assessment with heterogeneous data

Worn tools might lead to substantial detrimental implications on the surface integrity of workpieces for precision/ultra-precision machining. Most previous research has heavily relied on singular information, which might not be appropriate enough to ascertain tool conditions and guarantee the accuracy of workpieces. This paper proposes a CNN-LSTM hybrid model directly utilizing tool images to predict surface roughness on machined parts for tool condition assessment. This work first performs pruning based on UNet3+ architecture to eliminate redundant structures while integrating attention mechanisms to enhance the model's focus on the target region. On this basis, tool wear region information is intensely mined and heterogeneous data is optimized using Spearman correlation analysis. Subsequently, we innovatively proposed a hybrid model that integrates CNN and RNN, endowing the model with the ability to process spatial and sequential information. The effectiveness of the proposed methodology is validated using the practical data obtained from cutting experiments. The results indicate that the proposed tool condition assessment methodology significantly improves the segmentation accuracy of the tool wear region to 94.52 % (Dice coefficient) and predicts the surface roughness of machined parts with an accuracy exceeding 93.1 % (R2). It can be observed that the developed methodology may provide an effective solution for accurate tool condition assessment and the implementation of tool health management.

Effects of unilateral hippocampal surgical procedures needed for calcium imaging on mouse behavior and adult hippocampal neurogenesis

Hippocampus is essential for episodic memory formation, lesion studies demonstrating its role especially in processing spatial and temporal information. Further, adult hippocampal neurogenesis (AHN) in the dentate gyrus (DG) has also been linked to learning. To study hippocampal neuronal activity during events like learning, in vivo calcium imaging has become increasingly popular. It relies on the use of adeno-associated viral (AAV) vectors, which seem to lead to a decrease in AHN when applied on the DG. More notably, imaging requires the implantation of a relatively large lens into the tissue. Here, we examined how injection of an AAV vector and implantation of a 1-mm-diameter lens into the dorsal DG routinely used to image calcium activity impact the behavior of adult male C57BL/6 mice. To this aim, we conducted open-field, object-recognition and object-location tasks at baseline, after AAV vector injection, and after lens implantation. Finally, we determined AHN from hippocampal slices using a doublecortin-antibody. According to our results, the operations needed for in vivo imaging of the dorsal DG did not have adverse effects on behavior, although we noticed a decrease in AHN ipsilaterally to the operations. Thus, our results suggest that in vivo imaging can be safely used to, for example, correlate patterns of calcium activity with learned behavior. One should still keep in mind that the defects on the operated side might be functionally compensated by the (hippocampus in the) contralateral hemisphere.