Hierarchical Region Research Articles

Multi-scale hierarchical brain regions detect individual and interspecies variations of structural connectivity in macaque monkeys and humans

Macaques are representative animal models in translational research. However, the distinct shape and location of the brain regions between macaques and humans prevents us from comparing the brain structure directly. Here, we calculated structural connectivity (SC) with multi-scale hierarchical regions of interest (ROIs) to parcel out human and macaque brain into 8 (level 1 ROIs), 28 (level 2 ROIs), or 46 (level 3 ROIs) regions, which consist of anatomically and functionally defined level 4 ROIs (around 100 parcellation of the brain). The SC with the level 1 ROIs showed lower individual and interspecies variation in macaques and humans. SC with level 2 and 3 ROIs shows that the several regions in frontal, temporal and parietal lobe show distinct connectivity between macaques and humans. Lateral frontal cortex, motor cortex and auditory cortex were shown to be important areas for interspecies differences. These results provide insights to use macaques as animal models for translational study.

Incremental RPN: Hierarchical Region Proposal Network for Apple Leaf Disease Detection in Natural Environments.

Apple leaf diseases can seriously affect apple production and quality, and accurately detecting them can improve the efficiency of disease monitoring. Owing to the complex natural growth environment, apple leaf lesions may be easily confused with background noise, leading to poor performance. In this study, a cascaded Incremental Region Proposal Network (Inc-RPN) is proposed to accurately detect apple leaf diseases in natural environments. The proposed Inc-RPN has a two-layer RPN architecture, where the precursor RPN is leveraged to generate diseased leaf proposals, and the successor RPN focuses on extracting target disease spots based on diseased leaf proposals. In the successor RPN, a low-level feature aggregation module is designed to fully utilize the bridged features and preserve the semantic information of the target disease spots. An incremental module is also leveraged to extract aggregated diseased leaf features and target disease spot features. Finally, a novel position anchor generator is designed to generate anchors based on diseased leaf proposals. The experimental results show that the proposed Inc-RPN performs very well on the FALD_CED and Apple Leaf Disease datasets, showing that it can accurately perform apple leaf disease detection tasks.

Histopathology language-image representation learning for fine-grained digital pathology cross-modal retrieval

Large-scale digital whole slide image (WSI) datasets analysis have gained significant attention in computer-aided cancer diagnosis. Content-based histopathological image retrieval (CBHIR) is a technique that searches a large database for data samples matching input objects in both details and semantics, offering relevant diagnostic information to pathologists. However, the current methods are limited by the difficulty of gigapixels, the variable size of WSIs, and the dependence on manual annotations. In this work, we propose a novel histopathology language-image representation learning framework for fine-grained digital pathology cross-modal retrieval, which utilizes paired diagnosis reports to learn fine-grained semantics from the WSI. An anchor-based WSI encoder is built to extract hierarchical region features and a prompt-based text encoder is introduced to learn fine-grained semantics from the diagnosis reports. The proposed framework is trained with a multivariate cross-modal loss function to learn semantic information from the diagnosis report at both the instance level and region level. After training, it can perform four types of retrieval tasks based on the multi-modal database to support diagnostic requirements. We conducted experiments on an in-house dataset and a public dataset to evaluate the proposed method. Extensive experiments have demonstrated the effectiveness of the proposed method and its advantages to the present histopathology retrieval methods. The code is available at https://github.com/hudingyi/FGCR.

Learning to Aggregate Multi-Scale Context for Instance Segmentation in Remote Sensing Images.

The task of instance segmentation in remote sensing images, aiming at performing per-pixel labeling of objects at the instance level, is of great importance for various civil applications. Despite previous successes, most existing instance segmentation methods designed for natural images encounter sharp performance degradations when they are directly applied to top-view remote sensing images. Through careful analysis, we observe that the challenges mainly come from the lack of discriminative object features due to severe scale variations, low contrasts, and clustered distributions. In order to address these problems, a novel context aggregation network (CATNet) is proposed to improve the feature extraction process. The proposed model exploits three lightweight plug-and-play modules, namely, dense feature pyramid network (DenseFPN), spatial context pyramid (SCP), and hierarchical region of interest extractor (HRoIE), to aggregate global visual context at feature, spatial, and instance domains, respectively. DenseFPN is a multi-scale feature propagation module that establishes more flexible information flows by adopting interlevel residual connections, cross-level dense connections, and feature reweighting strategy. Leveraging the attention mechanism, SCP further augments the features by aggregating global spatial context into local regions. For each instance, HRoIE adaptively generates RoI features for different downstream tasks. Extensive evaluations of the proposed scheme on iSAID, DIOR, NWPU VHR-10, and HRSID datasets demonstrate that the proposed approach outperforms state-of-the-arts under similar computational costs. Source code and pretrained models are available at https://github.com/yeliudev/CATNet.

A Novel Image Segmentation Technique with N-L Means Filter Using Region Merging Strategy

Image segmentation is a branch of digital image processing which emphases on splitting an image into different fragments according to their features. This work proposes a novel image segmentation technique using fast non local (N-L) means filter with region merging approach. Here, a random noise is added to the original image and discrete wavelet transform (DWT) is applied to separate the low and high frequency bands of the digital image. The low frequency component is subjected to a fast N-L means filter and the filtered image is segmented by graph-based segmentation technique. Here, these segmented image regions are merged to reduce the over segmentation patterns. The detailed coefficients of the image component are treated with soft threshold filter to reduce the residual noise. Finally merged image is combined with preserved edge features at wavelet projection resulting in better segmented image. The results are compared with some of the best methods available in literature and are found to be significantly better. Keywords: Discrete wavelet transform (DWT), fast NL-means filter, graph-based segmentation, boundary based hierarchical region merging, soft-thresholding, peak signal to noise ratio (PSNR).

Accurate Prediction of Soil Heavy Metal Pollution Using an Improved Machine Learning Method: A Case Study in the Pearl River Delta, China.

In traditional soil heavy metal (HM) pollution assessment, spatial interpolation analysis is often carried out on the limited sampling points in the study area to get the overall status of heavy metal pollution. Unfortunately, in many machine learning spatial information enhancement algorithms, the additional spatial information introduced fails to reflect the hierarchical heterogeneity of the study area. Therefore, we designed hierarchical regionalization labels based on three interpolation techniques (inverse distance weight, ordinary kriging, and trend surface interpolation) as new spatial covariates for a machine learning (ML) model. It was demonstrated that regional spatial information improved the prediction performance of the model (R2 > 0.7). On the basis of the prediction results, the status of HM pollution in the Pearl River Delta (PRD) region was evaluated: cadmium (Cd) and copper (Cu) were the most serious pollutants in the PRD (the point overstandard rates are 18.77% and 12.95%, respectively). The analysis of index importance and bivariate local indicators of spatial association (LISA) shows that the key factors affecting the spatial distribution of heavy metals are geographical and climatic conditions [namely, altitude, humidity index, and normalized vegetation difference index (NDVI)] and some industrial activities (such as metal processing, printing and dyeing, and electronics industry). This study develops a novel approach to improve existing spatial interpolation techniques, which will enable more precise and scientific soil environmental management.

From center to surrounding: An interactive learning framework for hyperspectral image classification

Owing to rich spectral and spatial information, hyperspectral image (HSI) can be utilized for finely classifying different land covers. With the emergence of deep learning techniques, convolutional neural network (CNN), fully convolutional network (FCN), and recurrent neural network (RNN) have been widely applied in the field of HSI classification. Recently, transformer-based approaches represented by Vision Transformer (ViT) have yielded promising performance on numerous tasks and have been introduced to classify HSI. However, existing methods based on the above architectures still face three crucial issues that limit the classification performance: 1) geometric constraints caused by input data, 2) contribution fuzziness of central pixels with details, and 3) interaction gap between local areas and further environments. To tackle the above problems, an interactive learning framework inspired by ViT is proposed from a center to surrounding perspective, namely the center-to-surrounding interactive learning (CSIL) framework. Different from existing works, the CSIL framework enables to achieve multi-scale, detail-aware, and space-interactive classification based on a well-designed hierarchical region sampling strategy, center transformer, and surrounding transformer. Specifically, a hierarchical region sampling strategy is first proposed to flexibly generate the center region, neighbor region, and surrounding region, respectively. Thus, multi-scale input data breaks the geometric constraints. Second, a center transformer is presented to obtain core characteristics in detail based on the center region. In this way, central pixels are remarkably highlighted and the details are easily perceived. Third, a surrounding transformer including interactive self-attention learning is formulated for interacting both locally fine-grained distributions in the neighbor region and further coarse-grained environments in the surrounding region. With this structure, short- and long-term dependencies can be modeled, emphasized, and exchanged to bridge the interaction gap. Finally, the features from center transformer and surrounding transformer are integrated, then fed into a multi-layer perceptron for the optimization of semantic representation. Extensive experiments on six HSI datasets including small-, medium-, and large-scale scenes demonstrate the superiority over state-of-the-art CNN–, FCN-, RNN- and transformer-based approaches, even with very few training samples (for example 0.19% in complex HanChuan city scene). The source code will be available soon at https://github.com/jqyang22/CSIL.

Optimal Location of Hierarchical Regional Centers

人口減少が進む中，拠点の集約や交通ネットワークの再編に注目が集まっている．本研究ではコンパクト・プラス・ネットワークを念頭に置いた2階層の包含階層構造を持つ拠点を考え，両拠点とネットワークの運営費用及び拠点への移動費用を同時に最小化する数理モデルを構築する．近年，ビデオ通話等の普及を受け，上位の拠点に赴かなくとも居住地近隣の下位拠点で高次のサービスを受けられるようになると考えられる．そこで，構築したモデルを用いて需要規模が減少した場合及び下位拠点の需要捕捉率が上昇した場合の最適な拠点配置とネットワーク構造を計算し，今後の社会的変化に合わせて拠点配置とネットワーク構造を適切に変更させることが重要であることを明らかにした．

Camera and LiDAR-based point painted voxel region-based convolutional neural network for robust 3D object detection

Most of the three-dimensional (3D) object detection methods based on LiDAR point cloud data achieve relatively high performance in general cases. However, when the LiDAR points have noise or some corruptions, the detection performance can be severely affected. We propose a 3D object detection method that combines point cloud information with two-dimensional (2D) semantic segmentation information to enhance the feature representation for difficult cases, such as sparse, noisy, and partially absent data. Motivated by the Pointpainting techniques, we designed an early-stage fusion method based on a Voxel region-based convolutional neural network (R-CNN) architecture. The 2D semantic segmentation scores obtained by the Pointpainting techniques are appended to the raw point cloud data. The voxel-based features and 2D semantic information improve the performance in detecting instances when the point cloud is corrupted. In addition, we also designed a multiscale hierarchical region of interest pooling strategy that reduced the computational cost of Voxel R-CNN by at least 43%. Our method shows competitive results with the state-of-the-art methods on the standard KITTI dataset. In addition, three corrupted KITTI datasets, KITTI sparse (KITTI-S), KITTI jittering (KITTI-J), and KITTI dropout (KITTI-D), were used for robustness testing. With the noisy LiDAR points, our proposed point painted Voxel R-CNN achieved superior detection performance over that of the baseline Voxel R-CNN for the moderate case, with a notable improvement of 11.13% in average precision (AP) on the 3D object detection and 14.3% in AP on the bird’s eye view object detection.

Medical education, workforce inequalities, and hierarchical regionalism: the University of Lancaster and the unrealised medical school, 1964-68

Part of the justification for the Lancashire and South Cumbria New Hospitals Programme (NHP) is that a modern estateis necessary to attract high quality medical students as an international centre of excellence. Workforce proposals produced by Healthier Lancashire and South Cumbria (LSC) align this with plans for improved recruitment and retention of doctors. These also relate to efforts at reducing reliance on locums and improving patient access to modern specialist services. This interrelationship between the demands of medical education, workforce inequalities, and the organisation of services has a long history which precedes the current NHP proposals. Indeed, it has been a structural feature of the region’s National Health Service (NHS) since 1948. Using the attempt by the newly founded University of Lancaster to obtain a medical school from 1964 to 1968, this paper demonstrates the compounding relationship between medical education, workforce inequalities and hierarchical regionalism as a process of ‘sedimented governance’ which shapes the limits of future horizons for current health services leaders and the NHP.

Evaluation and optimal selection of Dunhuang radiometric calibration site based on OLI/Landsat 8 and MSI/Sentinel 2 data

ABSTRACT The Dunhuang radiometric calibration site is an important component of the China Radiometric Calibration Sites (CRCS), which has served for in-flight radiometric calibration of Chinese satellites sensors since the 1990s, whereby the homogeneity has been verified since its construction. However, due to the increasing demand for radiometric calibration of the in-flight satellite sensors, and the impacts of a changing natural environment and the destruction of human activities of the site, it is essential to reassess the homogeneity of Dunhuang site and redefine the suitable region for calibration. Accordingly, the homogeneity of the Dunhuang site is reevaluated based on the satellite images of OLI/Landsat 8 (with resolutions of 30 m) and MSI/Sentinel 2 (with resolutions of 10 m) observed from 2015 to 2021. An ROI of 50 × 75 km is selected covering three historical definitions of the Dunhuang site with an area of 25 × 25 km, 20 × 20 km, and 10 × 10 km, respectively. Firstly, the spatial-temporal coefficients of variation () of the ROI is calculated. The homogeneity of the three existing sites is analyzed, and the results show that the 20 km site (40.06° N−40.25° N, 94.25° E−94.42° E) has the best homogeneity among the existing sites in Dunhuang. Then, by using a hierarchical region statistical method, three optimal locations and scopes of the Dunhuang site (10 × 10 km: 40.044° N−40.144° N, 94.284° E−94.384° E; 20 × 20 km: 40.002° N−40. 202° N, 94.216° E−94.416° E; 25 × 25 km: 40.021° N−40.247° N, 94.071° E−94.364° E) are defined. The average of the three OGT sites decrease from about 0.94%～2% to 0.7%～1.54% relative to the existing sites, and the standard deviation of decrease from about 0.45%～2% to 0.24%～1%. The Getis statistics also indicate that they are more homogeneous than the existing definitions of the site. Additionally, the optimal Dunhuang sites are compared with the north Africa Pseudo-Invariant Calibration Sites (PICS). The homogeneity of optimal Dunhuang sites is proved to be better than most sites hence that could be the complement for each other. These three optimal Dunhuang sites are new candidate sites served for the satellite in-flight radiometric calibration.

A framework for spatial regionalization composed of novel clustering‐based algorithms under spatial contiguity constraints

AbstractTraditionally, the geospatial regionalization task consists of aggregating into regions, geographically connected areas that share similar characteristics. Although various spatial optimization approaches have been proposed for finding exact regionalization solutions, these approaches are not practical when applied to a large number of areas or problems for online aggregation, due to the long execution times using hardware with low resources. In this article, we present a framework for executing spatial regionalization tasks. The pre‐condition for using our framework is the definition of a map describing the neighborhood relations—as spatial contiguity constraints—among the areas (or objects) to be regionalized. In our framework we implemented three clustering algorithms with spatial contiguity constraints: RegK‐Means and Agglomerative Hierarchical Regionalization (AHR), our adaptation of k‐means partition‐based clustering and hierarchical‐based clustering algorithms, respectively; and the Automatic Zoning Procedure (AZP), a traditional algorithm for regionalization that also has the premise of simplifying the neighborhood relation representation. We conducted an exploratory analysis composed of two different experiments. Our results showed that our framework leads to a faster way of executing regionalization tasks in the experimental analysis, allowing us to observe a significant gain of AHR and RegK‐Means over AZP in execution time, while showing better or similar results in other metrics, such as figure of merit.

A hierarchical RCNN for vehicle and vehicle license plate detection and recognition

<span>Vehicle and vehicle license detection obtained incredible achievements during recent years that are also popularly used in real traffic scenarios, such as intelligent traffic monitoring systems, auto parking systems, and vehicle services. Computer vision attracted much attention in vehicle and vehicle license detection, benefit from image processing and machine learning technologies. However, the existing methods still have some issues with vehicle and vehicle license plate recognition, especially in a complex environment. In this paper, we propose a multivehicle detection and license plate recognition system based on a hierarchical region convolutional neural network (RCNN). Firstly, a higher level of RCNN is employed to extract vehicles from the original images or video frames. Secondly, the regions of the detected vehicles are input to a lower level (smaller) RCNN to detect the license plate. Thirdly, the detected license plate is split into single numbers. Finally, the individual numbers are recognized by an even smaller RCNN. The experiments on the real traffic database validated the proposed method. Compared with the commonly used all-in-one deep learning structure, the proposed hierarchical method deals with the license plate recognition task in multiple levels for sub-tasks, which enables the modification of network size and structure according to the complexity of sub-tasks. Therefore, the computation load is reduced.</span>

Hierarchical Region Learning and Patch Attention for Micro-Expression Recognition

Hierarchical Region Learning and Patch Attention for Micro-Expression Recognition

Local Scale-Guided Hierarchical Region Merging and Further Over- and Under-Segmentation Processing for Hybrid Remote Sensing Image Segmentation

With the development of medium- and high-resolution satellites, successfully segmenting differently sized geo-objects remains a challenging issue in the framework of geographic object-based image analysis (GEOBIA). The hybrid image segmentation method is a good alternative to produce good segmentation that best matched the different sizes of geo-objects. However, the existing methods almost use a certain scale or other segmentation parameters (SPs) to control the sizes and shapes of segments. This will lead to two issues: (1) one single scale is impossible to segment every geo-object well due to the land cover complexity within remote-sensing imageries; (2) over- and under-segmented regions still occur in the segmentation results, whatever using any advanced segmentation methods. To solve the above problems, this paper developed a hybrid image segmentation method with local scale-guided hierarchical region merging and further over- and under-segmentation processing. First, the primitive segmentation was produced and then stratified into layers with different land covers. Then, the local scale was calculated for a more objective merging process in the separating layers. Third, the over- and under-segmentation at separating layers was recognized and re-processed for achieving a fine segmentation. To validate the proposed method, it was applied to three test images of gaofen-1 satellite with different land cover types, and ten competing methods were compared. The visual and quantitative results indicated the advantage of our method in segmenting out different sizes of geo-objects, which can effectively reduce the over- and under-segmentation error.

Combining Static and Dynamic Load Balance in Parallel Routing for FPGAs

Routing is a very complex process in the field programmable gate array (FPGA) CAD flow. The increase of both FPGA size and design complexity leads to a long routing time hindering the productivity. In this article, we propose a more effective parallel router that combines static and dynamic load balance in parallel routing for FPGAs. First, we explore hierarchical region partitioning to assign routing tasks to different cores for static load balance. Then, we coordinate message propagation and task migration at runtime so that load balance between cores can be dynamically maintained in parallel routing. Finally, we combine static and dynamic load balance in the parallel routing for a higher degree of parallelism. Our parallel router performs on the multicore distributed-memory systems and the communication between cores is through message passing interface messages. We demonstrate the effectiveness of our parallel router using large-scale Titan designs. On average, our parallel router can scale up to 32 cores to achieve about 17× speedup with slight loss of quality, compared with the latest VTR 8 router.

Routing strategy of reducing energy consumption for underwater data collection

Underwater Wireless Sensor Networks (UWSNs) are widely used in many fields, such as regular marine monitoring and disaster warning. However, UWSNs are still subject to various limitations and challenges: ocean interferences and noises are high, bandwidths are narrow, and propagation delays are high. Sensor batteries have limited energy and are difficult to be replaced or recharged. Accordingly, the design of routing protocols is one of the solutions to these problems. Aiming at reducing and balancing network energy consumption and effectively extending the life cycle of UWSNs, this paper proposes a Hierarchical Adaptive Energy-efficient Clustering Routing (HAECR) strategy. First, this strategy divides hierarchical regions based on the depth of the sensor node in a three dimensional (3D) space. Second, sensor nodes form different competition radii based on their own relevant attributes and remaining energy. Nodes in the same layer compete freely to form clusters of different sizes. Finally, the transmission path between clusters is determined according to comprehensive factors, such as link quality, and then the optimal route is planned. The simulation experiment is conducted in the monitoring range of the 3D space. The simulation results prove that the HAECR clustering strategy is superior to LEACH and UCUBB in terms of balancing and reducing energy consumption, extending the network lifetime, and increasing the number of data transmissions.

The Olfactory Pathway Preserves Sensory Transmission During Sleep

Decreased responsiveness to sensory stimuli during sleep is thought to be mediated via a thalamic gate. However, olfactory information can gain access to olfactory cortices without an obligatory thalamic relay. It remains unknown how the olfaction system “gates” information flow during sleep. Here we simultaneously recorded local field potentials (LFPs) in hierarchical olfactory regions (olfactory bulb, piriform cortex, and orbitofrontal cortex) while optogenetically activating olfactory sensory neurons, ensuring consistent peripheral inputs across sleep/wake states in freely behaving mice. Surprisingly, evoked LFPs in sleep states (both NREM and REM) were larger in amplitude and contained greater gamma band power than in wakefulness. These findings argue against a central olfactory “gate,” further supported by single unit recordings in the piriform cortex showing nonattenuated information transmission during sleep. Quantitative analysis of nasal breathing revealed slower respiration rates and shallower inhalation slopes during sleep, supporting a peripheral “gate” that physically regulates state-dependent olfactory inputs.

Learning Methodologies to Generate Kernel-Learning-Based Image Downscaler for Arbitrary Scaling Factors

Displays and content have various resolutions and aspect ratios, requiring an image downscaler to adaptively reduce the image resolution. However, research on downscaling has garnered less attention than upscaling, including super-resolution. In practical display systems, simple interpolation, such as a bicubic filter that cannot preserve image details well, is still widely used for image downscaling rather than frame optimization-based or learning-based methods because of following reasons: frame optimization-based methods can effectively preserve image details after downscaling but are difficult to implement due to hardware costs. Learning-based methods have not been developed because defining a target downscaled image for training is difficult and training all downscaling factors is impossible. We propose a novel kernel-learning-based image downscaler to improve detail-preservation quality while supporting arbitrary downscaling factors using simple linear mapping. For this, a method to produce the ideal target downscaling result considering aliasing artifacts and detail preservation after downscaling is proposed. Then, we propose a training technique using the positional relationship between input and output pixels and a hierarchical region analysis to reproduce target images through simple kernel-based linear mapping. Lastly, a kernel-sharing technique is proposed to generate downscaling results for downscaling factors using a minimum number of trained kernels. In the simulation results, the proposed method demonstrated excellent edge preservation by improving the recall, precision, and F1 score, measuring the edge consistency between input and downscaled images, by up to 0.141, 0.079, 0.053, respectively, compared to benchmark methods. In a paired-comparison-based user study, the proposed method obtained the highest preference among benchmark methods using simple operations.

Der literarische Anschluss Siebenbürgens – Heinrich Zillich und Das Innere Reich (1934–1944): Eine Fallstudie

Abstract In the journal Das Innere Reich, published from 1934 to 1944 by the Langen-Müller publishing house in Munich, there is an astonishing density of articles relating to Transylvania. The south-eastern border region of the former Habsburg Monarchy is the subject of publications of essayistic contributions and texts by authors from the region, with Heinrich Zillich in particular playing a central role. The integration of Transylvania followed two basic lines: On the one hand, a group of people who had been living outside the inner-German language area for centuries was established as an integral part of the ʻGerman ethnic groupʼ. On the other hand, Transylvania, which has always been inhabited by different ethnic groups, was presented as a model of a clearly hierarchical region in the sense of the national ideology.