Natural Images Research Articles

Cross-Modal Contrastive Learning with a Style-Mixed Bridge for Single Image 3D Shape Retrieval

Image-based 3D shape retrieval (IBSR) is a cross-modal matching task which searches similar shapes from a 3D repository using a natural image. Continuous attention has been paid to this topic, such as joint embedding, adversarial learning, and contrastive learning. Modality gap and diversity of instance similarities are two obstacles for accurate and fine-grained cross-modal matching. To overcome the two obstacles, we propose a style-mixed contrastive learning method (SC-IBSR). On one hand, we propose a style transition module to mix the styles of images and rendered shape views to form an intermediate style and inject it into image contents. The obtained style-mixed image features serve as a bridge for later contrastive learning in order to alleviate the modality gap. On the other hand, the proposed strategy of fine-grained consistency constraint aims at cross-domain contrast and considers the different importance of negative (positive) samples. Extensive experiments demonstrate the superiority of the style-mixed cross-modal contrastive learning on both the instance-level retrieval benchmark (i.e., Pix3D, Stanford Cars, and Comp Cars that annotate shapes to images), and the unsupervised category-level retrieval benchmark (i.e., MI3DOR-1 and MI3DOR-2 with unlabeled 3D shapes). Moreover, experiments are conducted on Office-31 dataset to validate the generalization capability of our method. Code and pretrained models will be available at https://github.com/honoria0204/SC-IBSR .

QAGA-Net: enhanced vision transformer-based object detection for remote sensing images

PurposeVision transformers (ViT) detectors excel in processing natural images. However, when processing remote sensing images (RSIs), ViT methods generally exhibit inferior accuracy compared to approaches based on convolutional neural networks (CNNs). Recently, researchers have proposed various structural optimization strategies to enhance the performance of ViT detectors, but the progress has been insignificant. We contend that the frequent scarcity of RSI samples is the primary cause of this problem, and model modifications alone cannot solve it.Design/methodology/approachTo address this, we introduce a faster RCNN-based approach, termed QAGA-Net, which significantly enhances the performance of ViT detectors in RSI recognition. Initially, we propose a novel quantitative augmentation learning (QAL) strategy to address the sparse data distribution in RSIs. This strategy is integrated as the QAL module, a plug-and-play component active exclusively during the model’s training phase. Subsequently, we enhanced the feature pyramid network (FPN) by introducing two efficient modules: a global attention (GA) module to model long-range feature dependencies and enhance multi-scale information fusion, and an efficient pooling (EP) module to optimize the model’s capability to understand both high and low frequency information. Importantly, QAGA-Net has a compact model size and achieves a balance between computational efficiency and accuracy.FindingsWe verified the performance of QAGA-Net by using two different efficient ViT models as the detector’s backbone. Extensive experiments on the NWPU-10 and DIOR20 datasets demonstrate that QAGA-Net achieves superior accuracy compared to 23 other ViT or CNN methods in the literature. Specifically, QAGA-Net shows an increase in mAP by 2.1% or 2.6% on the challenging DIOR20 dataset when compared to the top-ranked CNN or ViT detectors, respectively.Originality/valueThis paper highlights the impact of sparse data distribution on ViT detection performance. To address this, we introduce a fundamentally data-driven approach: the QAL module. Additionally, we introduced two efficient modules to enhance the performance of FPN. More importantly, our strategy has the potential to collaborate with other ViT detectors, as the proposed method does not require any structural modifications to the ViT backbone.

Abstract 4113716: Natural shear wave imaging as a predictor for left ventricular pressures: Can they reflect raised filling pressures?

Background: Current conventional echocardiographic assessment of diastolic function relies on predicting left ventricular (LV) filling pressure as a surrogate of diastolic function. High frame rate (HFR) Shear wave (SW) imaging is emerging as an innovative parameter for assessing myocardial stiffness, a key determinant of diastolic function. Natural SW are induced by valve closure (i.e. mitral valve closure (MVC)). Little is known so far, to which extent SWs could reflect changes in diastolic LV filling pressures. Purpose: We aimed at investigating the relationship between natural SW after MVC and LV diastolic pressures in an attempt to set up a cut-off value beyond which, elevated LV pressure can be suspected indicating diastolic dysfunction. Methods: Thirty patients (mean age 68.4±9.9 years) scheduled for clinically indicated left heart catheterization were prospectively recruited. Left ventricular end-diastolic pressure (LVEDP) was measured during the catheterization. Immediately afterwards, conventional as well as HFR (1167 ±86 Hz) echocardiography was performed. We drew an anatomical M-mode along the anteroseptal wall in the parasternal long axis view from base to apex and the display was colour coded for tissue acceleration. SWs appeared immediately after MVC as tilted green bands and their propagation velocity was measured semi-automatically (Figure 1) . Results: LVEDP in our cohort ranged from 7.5 to 29.5 mmHg. From conventional echocardiographic parameters, only mitral inflow to mitral relaxation velocity ratio (E/e’) correlated moderately with LVEDP (r=0.51, p=0.006). SW velocities after MVC showed strong positive correlation with LVEDP (r=0.78, p<0.001) (Figure 2) . A cut-off value for SW velocity of 5.8 m/s showed an excellent ability to discriminate normal from elevated LVEDP (AUC: 0.98; Sensitivity: 90%, specificity: 100%, accuracy: 97%) (Figure 3) . Conclusions: Our data showed that SW velocities at end-diastole after MVC could predict elevated LV filling pressures accurately. Natural shear wave imaging appears as a promising innovative tool for assessment of diastolic function.

A Low-Cost Remotely Configurable Electronic Trap for Insect Pest Dataset Generation

The precise monitoring of insect pest populations is the foundation of Integrated Pest Management (IPM) for pests of plants, humans, and animals. Digital technologies can be employed to address one of the main challenges, such as reducing the IPM workload and enhancing decision-making accuracy. In this study, digital technologies are used to deploy an automated trap for capturing images of insects and generating centralized repositories on a server. Subsequently, advanced computational models can be applied to analyze the collected data. The study provides a detailed description of the prototype, designed with a particular focus on its remote reconfigurability to optimize repository quality; and the server, accessible via an API interface to enhance system interoperability and scalability. Quality metrics are presented through an experimental study conducted on the constructed demonstrator, emphasizing trap reliability, stability, performance, and energy consumption, along with an objective analysis of image quality using metrics such as RMS contrast, Image Entropy, Image sharpness metric, Natural Image Quality Evaluator (NIQE), and Modulation Transfer Function (MFT). This study contributes to the optimization of the current knowledge regarding automated insect pest monitoring techniques and offers advanced solutions for the current systems.

V1 neurons are tuned to perceptual borders in natural scenes

The visual system needs to identify perceptually relevant borders to segment complex natural scenes. The primary visual cortex (V1) is thought to extract local borders, and higher visual areas are thought to identify the perceptually relevant borders between objects and the background. To test this conjecture, we used natural images that had been annotated by human observers who marked the perceptually relevant borders. We assessed the effect of perceptual relevance on V1 responses using human neuroimaging, macaque electrophysiology, and computational modeling. We report that perceptually relevant borders elicit stronger responses in the early visual cortex than irrelevant ones, even if simple features, such as contrast and the energy of oriented filters, are matched. Moreover, V1 neurons discriminate perceptually relevant borders surprisingly fast, during the early feedforward-driven activity at a latency of ~50 ms, indicating that they are tuned to the features that characterize them. We also revealed a delayed, contextual effect that enhances the V1 responses that are elicited by perceptually relevant borders at a longer latency. Our results reveal multiple mechanisms that allow V1 neurons to infer the layout of objects in natural images.

Semantic Guided Matting Net

Abstract Human matting refers to extracting human parts from natural images with high quality, including human detail information such as hair, glasses, hats, etc. This technology plays an essential role in image synthesis and visual effects in the film industry. When the green screen is not available, the existing human matting methods need the help of additional inputs (such as trimap, background image, etc.), or the model with high computational cost and complex network structure, which brings great difficulties to the application of human matting in practice. To alleviate such problems, we use a segmentation network as the foundation and use multiple branches to achieve human segmentation, contour detail extraction, and information fusion. We also propose a foreground probability map module, which uses the feature maps in the segmentation network to pre-estimate the foreground probabilities of each pixel and obtain Semantic Guided Matting Net. Under the condition that only a single image is needed as the input, the human matting task can be realized by making full use of the semantic information in the image. We validate our method on the P3M-10k dataset. Compared with the benchmark, our method has made significant improvements in various evaluation indicators.

Deep Learning-Based Pulmonary Nodule Screening: A Narrative Review

AbstractGiven its capacity to generate three-dimensional pictures, computed tomography is the most effective means of detecting lung nodules with more excellent resolution of detected nodules. Small lung nodules can easily be overlooked on chest X-rays, making interpretation difficult. Artificial intelligence algorithms have recently demonstrated remarkable progress in medical imaging, especially with deep learning techniques such as convolutional neural networks (CNNs). CNN produces excellent results in natural image recognition and classification using abundant available data and the computational abilities of modern computers. It further reduces false-positive pulmonary nodules in medical image processing. This review article provides a detailed and inclusive review of recent advances, challenges, performance comparisons, and possible future directions for the problem of pulmonary nodule screening using deep learning methods.

Szakrális kéziratok szövegeinek szimbolikája

Symbols have always been an integral and permanent part of our poetry. Although their presence may be fading, they will never completely disappear. Symbols simultaneously influence both our conscious and subconscious thought, providing interpretive variations to encode meaning. A single word can often encapsulate an entire existential feeling that we might not be able to express or describe more clearly and emotionally, even with the help of a paraphrase. Of course, there are times when our emotional or even social circumstances prevent us from making specific declarations, manifestations. This study examines the textual symbolism of a sanctified manuscript collected in Solovka, a village located in Transcarpathia. The research’s mentioned source area is a settlement that has received little attention in the academic literature. The study offers insight into the traditional spirit that is still evident in the religious communities of the settlement today. It provides an opportunity for a deeper, reinterpreted understanding of the texts preserved by the community and their significance. We are working at the intersection of folklore and literature. The study reveals, among other things, that there is a strong connection between the two areas. During the analysis a number of symbols were identified that carry deeper meanings, especially in natural images. We dealt separately with the meanings of the flower, the tree, the torch, the seasons and the coffin. As an example, we compared the works of lyric poets with “folk writings”, focusing on the use of symbols. During the examination of texts intended for analysis, a clear parallel between ethnography and literature became evident. This connection was demonstrated through a comparison of certain literary works, poems and texts of folk manuscripts. Poets such as Friedrich Schiller, János Arany, Endre Ady, Dezső Kosztolányi, Mihály Csokonai Vitéz, and Sándor Weöres were quoted. The study also includes comparisons of the symbolism found in texts of wedding and funeral ceremonies. Despite the significant emotional distance between these stages of human life, text analysis suggests many similarities.

Are Australian Food Companies Using Nature Imagery as a Marketing Strategy? A Content Analysis

Emerging evidence suggests that nature imagery can affect consumer food choices. However, little is known about whether and, if so, how food companies use nature imagery as a means to appeal to consumers’ environmental concerns as a marketing strategy. The current study aims to explore this matter for websites of Australian food companies. We extracted images from 45 Australian food company websites and coded for (1) nature type, (2) restorative effect, (3) level of cultivation and human interference, and (4) processing level (only for images with food present). All included food companies had a separate webpage dedicated to sustainability, and most nature images were located there. Sustainability pages had more images of ‘pure nature’ (i.e., no cultivation or human interference and high in restorative effects) and more ‘urban’ images compared to the main pages. Most food in the images was unprocessed, while only one image featured both nature and food. Australian food companies are commonly using images of nature on their sustainability webpages. The images might be used as a marketing strategy to connect with consumers’ environmental concerns as a strategy to build trust and affinity with customers. However, companies currently do not combine nature imagery and their food products in one image on their websites.

[formula omitted]-norm regularized quaternion matrix completion using sparse representation and approximate QSVD

Color image completion is a challenging problem in computer vision. Despite the effectiveness of low-rank quaternion matrix completion (LRQMC) methods, their reliance on quaternion singular value decompositions (QSVD) makes them computationally expensive. We propose a novel method based on quaternion Qatar Riyal decomposition (QQR) and quaternion L2,1-norm, named QLNM-QQR, which reduces computational complexity by avoiding the need for calculating the QSVD of large quaternion matrices. Furthermore, we introduce two improvements: IRQLNM-QQR, which utilizes iteratively reweighted quaternion L2,1-norm minimization, and QLNM-QQR-SR, which integrates sparse regularization. Additionally, an analysis of the convergence of the proposed algorithms is conducted. Our experiments show that IRQLNM-QQR outperforms QLNM-QQR on both natural and medical color images. The QLNM-QQR-SR method achieves higher PSNR and SSIM values than state-of-the-art methods, with PSNR averaging 4-8 dB higher than non-sparse methods and 0.4 dB higher than the sparse-based TNN-SR method.

A histogram-based approach to calculate graph similarity using graph neural networks

Deep learning has revolutionized the field of pattern recognition and machine learning by exhibiting exceptional efficiency in recognizing patterns. The success of deep learning can be seen in a wide range of applications including speech recognition, natural language processing, video processing, and image classification. Moreover, it has also been successful in recognizing structural patterns, such as graphs. Graph Neural Networks (GNNs) are models that employ message passing between nodes in a graph to capture its dependencies. These networks memorize a state that approximates graph information with greater depth compared to traditional neural networks. Although training a GNN can be challenging, recent advances in GNN variants, including Graph Convolutional Neural Networks, Gated Graph Neural Networks, and Graph Attention Networks, have shown promising results in solving various problems. In this work, we present a GNN-based approach for computing graph similarity and demonstrate its application to a classification problem. Our proposed method converts the similarity of two graphs into a score, and experiments on state-of-the-art datasets show that the proposed technique is effective and efficient. Results are summarized using a confusion matrix and mean square error metric, demonstrating the accuracy of our proposed technique.

The Short Video Platform Recommendation Mechanism Based on the Improved Neural Network Algorithm to the Mainstream Media

Currently, the capacity of short videos continues to increase. Video manufacturers hope to enhance user experience and stickiness through recommendation algorithms, while users seek personalized videos to save time and money. Therefore, in order to address the data sparsity and high-dimensional feature extraction, this study proposes a novel short video platform recommendation model. The proposed method utilizes the term frequency inverse document frequency algorithm for text mining, and combines error back propagation neural network for learning to explore the potential connection between users and videos. This method combines natural language processing and image analysis in deep learning to construct accurate user and video models, deeply explore user interests, and improve the accuracy and effectiveness of recommendation systems for user preferences. The research results showed that the recommendation accuracy of this method was 66% and 70% respectively, and the prediction accuracy was 73.50% and 88% respectively. When Num=128, 200 data points were recommended within 0.3678s. The proposed algorithm outperforms the other three methods in terms of recommendation accuracy compared with the ItemCF and UserCF algorithms. This is because the method uses an approach based on image and user vector, combined with relevant features of the video and user. The proposed method can deeply explore the relevant features of videos and users, overcoming the data scarcity in previous collaborative screening, and guiding video recommendation on practical media platforms.

Unifying domain gap in Federated Learning: A geometric approach

Federated Learning (FL) is a framework designed to enable distributed and privacy-preserving learning. It typically involves a central server coordinating multiple clients to collaboratively train a global model without the need for sharing private data. Recent studies have revealed that this parameter-only communication strategy can be hindered by non-independent and identically distributed (non-iid) data across clients, leading to statistical heterogeneity. This issue necessitates the development of Personalized Federated Learning (PFL), which aims to enhance the performance of individual clients while still allowing for the learning of a global model. Most existing work on PFL has focused on label shift in supervised tasks, with relatively little attention given to feature shift, another common problem in real-world applications of federated learning. For instance, client data may vary in terms of illumination, angles, and quality for natural images, or in imaging protocols for medical images. To address this issue, we propose a general yet powerful framework that leverages the geometric interpretation of deep neural networks. Geometrically, a deep neural network partitions the feature space into a Power Diagram (PD) or Voronoi Diagram (VD). Regularized losses are introduced to mitigate feature shift from the perspective of PD and VD. Additionally, a personalized generator network is formulated for each client to rectify feature distribution and adapt the input data from its own domain to an ”averaged” or unified domain that generalizes well across all clients. Our method demonstrates competitive performance compared to current state-of-the-art methods.

Adversarial attacks in signature verification: a deep learning approach

Handwritten signature recognition in forensic science is crucial for identity and document authentication. While serving as a legal representation of a person’s agreement or consent to the contents of a document, handwritten signatures de termine the authenticity of a document, identify forgeries, pinpoint the suspects and support other pieces of evidence like ink or document analysis. This work focuses on developing and evaluating a handwritten signature verification sys tem using a convolutional neural network (CNN) and emphasising the model’s efficacy using hand-crafted adversarial attacks. Initially, handwritten signatures have been collected from sixteen volunteers, each contributing ten samples, fol lowed by image normalization and augmentation to boost synthetic data samples and overcome the data scarcity. The proposed model achieved a testing accu racy of 91.35% using an 80:20 train-test split. Additionally, using the five-fold cross-validation, the model achieved a robust validation accuracy of nearly 98%. Finally, the introduction of manually constructed adversarial assaults on the sig nature images undermines the model’s accuracy, bringing the accuracy down to nearly 80%. This highlights the need to consider adversarial resilience while designing deep learning models for classification tasks. Exposing the model to real look-alike fake samples is critical while testing its robustness and refining the model using trial and error methods.

Simultaneous image denoising and completion through convolutional sparse representation and nonlocal self-similarity

Low rank matrix approximation (LRMA) has been widely studied due to its capability of approximating original image from the degraded image. According to the characteristics of degraded images, image denoising and image completion have become research objects. Existing methods are usually designed for a single task. In this paper, focusing on the task of simultaneous image denoising and completion, we propose a weighted low rank sparse representation model and the corresponding efficient algorithm based on LRMA. The proposed method integrates convolutional analysis sparse representation (ASR) and nonlocal statistical modeling to maintain local smoothness and nonlocal self-similarity (NLSM) of natural images. More importantly, we explore the alternating direction method of multipliers (ADMM) to solve the above inverse problem efficiently due to the complexity of simultaneous image denoising and completion. We conduct experiments on image completion for partial random samples and mask removal with different noise levels. Extensive experiments on four datasets, i.e., Set12, Kodak, McMaster, and CBSD68, show that the proposed method prevents the transmission of noise while completing images and has achieved better quantitative results and human visual quality compared to 17 methods. The proposed method achieves (1.9%, 1.8%, 4.2%, and 3.7%) gains in average PSNR and (4.2%, 2.9%, 6.7%, and 6.6%) gains in average SSIM over the sub-optimal method across the four datasets, respectively. We also demonstrate that our method can handle the challenging scenarios well. Source code is available at https://github.com/weimin581/demo_CSRNS.

Study on Eyebrow Shape and Image Suitable for Men’s Social Image -Focusing on South Korean and Chinese Men in their 20’s and 30’s-

This paper aimed to run a comparative analysis of preferred eyebrow shapes among South Korean and Chinese men in their 20’s and 30’s, to present their desirable eyebrow shape effective for giving an impression. Furthermore, the renewed recognition of the social function of eyebrow shapes, an important factor which determines their impression formation depending upon desirable images, could help ramp up the related cosmetics industry. For this purpose, we carried out a survey and did statistical analysis. The findings of this paper were as follows. First, as for preferred eyebrow shape, there turned out a common predominant preference for ‘standard type’ among South Korean and Chinese men. However, South Korean men had the least preference for ‘angled type’, while Chinese men had the least preference for ‘straight type’. This means Chinese men go after a ‘masculine’ image, compared with South Korean men. Second, among both South Korean and Chinese men, a ‘natural’ image was commonly the most popular, with ‘cute’ image the least popular. However, ‘simple’, ‘tender’, ‘refined’, and ‘urbane’ images ranked high in popularity among South Korean men; on the other hand, Chinese men had a preference for ‘masculine’, ‘simple’, ‘decent’, ‘dynamic’, and ‘tough’ images. Third, among men with sales/service or self-employed jobs which entail face to face contact with people, a ‘natural’ image without revealing their personality was commonly most popular; on the other hand, there was a tendency to raise their energy through ‘dynamic’ images among men with office/administrative jobs which need human resource management through work concentration and hierarchy. And men with production/engineering jobs which entail lots of movements tended to seek toughness through a ‘masculine’ image. There was a preference for bright and affirmative energy through ‘lively’ image among students who was building relationships with friends with no occupational image. Finally, as regards socially desirable eyebrow shapes of those surveyed by job, ‘standard type’ or ‘straight type’ is suitable for men with sales/service or self-employed jobs requiring a ‘natural image’. On the other hand, ‘upward type’ is suitable for men with production/engineering job in need of a ‘masculine image’. And also, ‘angled type’ goes well with a ‘dynamic image’ which men with office/administrative job need.

Super-resolution imaging quality enhancement method for distributed array infrared camera

Abstract To address issues related to low resolution and high noise in infrared cameras, a distributed array infrared camera imaging system utilizing four cameras is proposed. The four cameras are arranged in an unconstrained array, and the combination algorithm of Projections onto Convex Sets (POCS) and Real-Enhanced Super-Resolution Generative Adversarial Networks (Real-ESRGAN) is applied to achieve high-quality super-resolution infrared imaging. The wavelet fusion algorithm is used to preprocess four low-resolution infrared images to reduce noise. Then, the POCS algorithm is used to reconstruct the preprocessed image. Finally, the Real-ESRGAN is employed for image reconstruction, resulting in an ultra-high-resolution infrared image. The results show that compared to single infrared camera imaging, the resolution of images reconstructed using the distributed infrared camera array is increased by 0.58 times, with the Modulation Transfer Function (MTF) increased by 1.2 times. Additionally, the entropy is increased by 18.87%, the standard deviation is increased by 13.51%, and the Naturalness Image Quality Evaluator (NIQE) is reduced by 18.87%. This demonstrates a significant enhancement in the super-resolution imaging quality of the distributed infrared camera array.

Integration of geoelectric methods and remote sensing in assessing the environment state in the area of impact of ecologically hazardous objects

Relevance. The need for comprehensive assessment of environmental components in the zones of ecologically hazardous industrial facilities impact, which enables to operatively identify the spatial structure of technogenic pollution. Aim. Chemical pollution assessment of the geological environment in the zones of industrial objects impact, based on the combined use of remote sensing and ground-based geoelectric methods. Objects. Sources of pollution during the production of mineral fertilizers (phosphogypsum dumps, sulfur storage). Methods. The author has developed a complex of methods, consisting of multispectral space imaging (Sentinel-2 satellite images) and ground-based geoelectric researches: water surface resistivity measurements, natural electric field potential imaging, vertical electrical sounding. According to satellite image data (vegetation indices NDVI, NBR, SWVI), technogenic geosystems modifications are distinguished, characterizing different levels of chemical pollution (TM-1, TM-2 and natural geosystem). Within the modification zones, geoelectrical studies are carried out to assess the depth of contamination. Results. In the zone of phosphogypsum dumps impact, two technogenic modifications were distinguished, with a total area of 83.3 hectares, characterized by a statistically significant decrease in the values of vegetation indices compared to the average level. Based on the results of vertical electrical sounding, it was established that the depth of contamination in the TM-1 zone reaches up to 20 m, in the TM-2 zone – up to 2.5 m. The migration of contaminated substances from the dumps occurs with surface drain. In the zone of impact of the sulfur storage, the area of pollution, specified with vegetation indices, covers two hectares. According to the results of the natural electric field method and vertical electrical sounding, the flow of pollution from the sulfur storage site penetrates into groundwater, which is discharged in the TM-1 zone, causing the death of vegetation. In the TM-2 zone, contamination is registered only in the ground aquifer. The developed set of methods enables to quickly and relatively comprehensively diagnose the state of the undergoing technogenic impacts geological environment.

Research on the Registration of Aerial Images of Cyclobalanopsis Natural Forest Based on Optimized Fast Sample Consensus Point Matching with SIFT Features

The effective management and conservation of forest resources hinge on accurate monitoring. Nonetheless, individual remote-sensing images captured by low-altitude unmanned aerial vehicles (UAVs) fail to encapsulate the entirety of a forest’s characteristics. The application of image-stitching technology to high-resolution drone imagery facilitates a prompt evaluation of forest resources, encompassing quantity, quality, and spatial distribution. This study introduces an improved SIFT algorithm designed to tackle the challenges of low matching rates and prolonged registration times encountered with forest images characterized by dense textures. By implementing the SIFT-OCT (SIFT omitting the initial scale space) approach, the algorithm bypasses the initial scale space, thereby reducing the number of ineffective feature points and augmenting processing efficiency. To bolster the SIFT algorithm’s resilience against rotation and illumination variations, and to furnish supplementary information for registration even when fewer valid feature points are available, a gradient location and orientation histogram (GLOH) descriptor is integrated. For feature matching, the more computationally efficient Manhattan distance is utilized to filter feature points, which further optimizes efficiency. The fast sample consensus (FSC) algorithm is then applied to remove mismatched point pairs, thus refining registration accuracy. This research also investigates the influence of vegetation coverage and image overlap rates on the algorithm’s efficacy, using five sets of Cyclobalanopsis natural forest images. Experimental outcomes reveal that the proposed method significantly reduces registration time by an average of 3.66 times compared to that of SIFT, 1.71 times compared to that of SIFT-OCT, 5.67 times compared to that of PSO-SIFT, and 3.42 times compared to that of KAZE, demonstrating its superior performance.

Flexible Natural Language-Based Image Data Downlink Prioritization for Nanosatellites

Nanosatellites increasingly produce more data than can be downlinked within a reasonable time due to their limited bandwidth and power. Therefore, an on-board system is required to prioritize scientifically significant data for downlinking, as described by scientists. This paper determines whether natural language processing can be used to prioritize remote sensing images on CubeSats with more flexibility compared to existing methods. Two approaches implementing the same conceptual prioritization pipeline are compared. The first uses YOLOv8 and Llama2 to extract image features and compare them with text descriptions via cosine similarity. The second approach employs CLIP, fine-tuned on remote sensing data, to achieve the same. Both approaches are evaluated on real nanosatellite hardware, the VERTECS Camera Control Board. The CLIP approach, particularly the ResNet50-based model, shows the best performance in prioritizing and sequencing remote sensing images. This paper demonstrates that on-orbit prioritization using natural language descriptions is viable and allows for more flexibility than existing methods.