Low Resolution Research Articles

Heavy black hole seed formation in high-z atomic cooling halos

Context. Halos with masses in excess of the atomic limit are believed to be ideal environments in which to form heavy black hole seeds with masses above 103 M⊙. In cases where the H2 fraction is suppressed, this is expected to lead to reduced fragmentation of the gas and the generation of a top-heavy initial mass function. In extreme cases this can result in the formation of massive black hole seeds. Resolving the initial fragmentation scale and the resulting protostellar masses has, until now, not been robustly tested. Aims. We run zoom-in simulations of atomically cooled halos in which the formation of H2 is suppressed to assess whether they can truly resist fragmentation at high densities and tilt the initial mass function towards a more top-heavy form and the formation of massive black hole seeds. Methods. Cosmological simulations were performed with the moving mesh code AREPO, using a primordial chemistry network until z ∼ 11. Three haloes with masses in excess of the atomic cooling mass were then selected for detailed examination via zoom-ins. A series of zoom-in simulations, with varying levels of maximum spatial resolution, captured the resulting fragmentation and formation of metal-free stars using the sink particle technique. The highest resolution simulations resolved densities up to 10−6 g cm−3 (1018 cm−3) and captured a further 100 yr of fragmentation behaviour at the centre of the halo. Lower resolution simulations were then used to model the future accretion behaviour of the sinks over longer timescales. Results. Our simulations show intense fragmentation in the central region of the halos, leading to a large number of near-solar mass protostars. Even in the presence of a super-critical Lyman-Werner radiation field (JLW > 105J21), H2 continues to form within the inner ∼2000 au of the halo. Despite the increased fragmentation, the halos produce a protostellar mass spectrum that peaks at higher masses relative to standard Population III star-forming halos. The most massive protostars have accretion rates of 10−3–10−1 M⊙ yr−1 after the first 100 years of evolution, while the total mass of the central region grows at 1 M⊙ yr−1. Lower resolution zoom-ins show that the total mass of the system continues to accrete at ∼1 M⊙ yr−1 for at least 104 yr, although how this mass is distributed amongst the rapidly growing number of protostars is unclear. However, assuming that a fraction of stars can continue to accrete rapidly, the formation of a sub-population of stars with masses in excess of 103 M⊙ is likely in these halos. In the most optimistic case, we predict the formation of heavy black hole seeds with masses in excess of 104 M⊙, assuming an accretion behaviour in line with expectations from super-competitive accretion and/or frequent mergers with secondary protostars.

Measurement of mandibular incisive canal diameter using cone beam computed tomography

Cone beam computed tomography was used for measuring the diameter of a Mandibular incisive canal. The dental practice can now visualize small bone structures with low dose and high spatial resolution due to the introduction of CBCT (Cone Beam Computed Tomography). Therefore, it is of interest to measure the diameter of the MIC (Mandibular Incisive Canal) using the CBCT. Fifty CBCT scans obtained for implant placement preoperative planning were included in the study material. Observers carefully examined the CBCT data to determine the MIC’s presence & position. The inner diameter of the MIC was measured by taking the longest distance between the inner cortical borders of the canal which was further analyzed using software CS3D software. The collected data were further subjected to statistical analysis. The average mean of the population was 1.7130±.5178. The average diameter of MIC in males was 1.735±.5406 and in females was 1.6800±.4934. Complications from implant surgery in the anterior mandible may be prevented by utilizing CBCT scans to analyze the smallest diameter of the MIC.

SOHD-LOAM: Suppress odometry height drift lidar odometry and mapping on undulating road

In the process of mapping outdoor undulating and flat roads, existing LiDAR SLAM systems often encounter issues such as map distortion and ghosting. These problems arise due to the low vertical resolution of multi-line LiDAR, which easily leads to the occurrence of odometry height drift during the mapping process. To address this challenge, this study propose a novel LiDAR SLAM system named SOHD-LOAM, designed specifically to suppress odometry height drift. This system encompasses several critical components, including data preprocessing, front-end LiDAR odometry, back-end LiDAR mapping, loop detection, and graph optimization. SOHD-LOAM leverages the road gradient limitation algorithm and the height smoothing algorithm as its core, while also integrating the Kalman filter, loop detection, and graph optimization techniques. To evaluate the performance of SOHD-LOAM, the comprehensive experiments are conducted with using KITTI datasets and real-world scenes. The experimental results demonstrate that SOHD-LOAM achieves superior accuracy and robustness in global odometry compared to the state-of-the-art LEGO-LOAM. Specifically, the height error of the sequences 00, 05 experiment was found to be 40.62% and 61.92% lower than that of LEGO-LOAM. Additionally, the maps generated by SOHD-LOAM exhibit no distortion or ghosting, thereby significantly enhancing map quality.

Dual-Mode Solidly Mounted Resonator-Based Sensor for Temperature and Humidity Detection and Discrimination.

Sensors based on solidly mounted resonators (SMRs) exhibit a good set of properties, such as high sensitivity, fast response, low resolution limit and low production cost, which makes them an appealing technology for sensing applications. However, they can suffer from cross-sensitivity issues, as their response can be altered by undesirable ambient factors, such as temperature and humidity variations. In this work we propose a method to discriminate humidity variations from the general frequency response using an SMR specifically manufactured to operate in a dual-mode (displaying two close resonances). The two modes behave similarly towards humidity changes (-1.94 kHZ/(%RH)) for resonance one and -1.62 kHZ/(%RH) for resonance two), whereas their performance under temperature changes is significantly different, displaying 2.64 kHZ/°C for resonance one and 34.21 kHZ/°C for resonance two. This allows for the decoupling process to be carried out in a straightforward manner. Frequency response is tracked under different humidity conditions, in the -20 °C to room temperature region, proving that this behavior is reproducible in any given environment.

Enhancement of X-Ray Images by Gradient Domain Guided Filtering and Non Sub Sampled Shearlet Transform

Abstract: In order to address the issues of low resolution, noise amplification, missing details, and weak edge gradient retention in the X-ray image improvement process, we propose in this study an image enhancement technique that combines nonSubsampled Shearlet transform with gradient-domain guided filtering. We start by breaking down the non Subsampled Shearlet transform and histogram equalization into the original image. We receive multiple high-frequency sub-bands as well as a lowfrequency sub-band. To enhance the overall contrast of the image and bring out the contour information in the low-frequency sub-band, adaptive gamma correction with weighting distribution is applied. To reduce image noise and emphasize edge information and detail, gradient-domain guided filtering is applied to the high-frequency sub-bands. In order to generate the final enhanced image, we use the inverse non-Subsampled Shearlet transform to reconstruct all of the sub-bands that were successfully processed. The experimental findings demonstrate the effectiveness of the suggested algorithm in improving X-ray images, as well as the clear advantages its objective index provides over certain conventional algorithms.

3D-Printed Conformal Meta-Lens with Multiple Beam-Shaping Functionalities for Mm-Wave Sensing Applications.

In this paper, a 3D conformal meta-lens designed for manipulating electromagnetic beams via height-to-phase control is proposed. The structure consists of a 40 × 20 array of tunable unit cells fabricated using 3D printing, enabling full 360° phase compensation. A novel automatic synthesizing method (ASM) with an integrated optimization process based on genetic algorithm (GA) is adopted here to create the meta-lens. Simulation using CST Microwave Studio and MATLAB reveals the antenna's beam deflection capability by adjusting phase compensations for each unit cell. Various beam scanning techniques are demonstrated, including single-beam, dual-beam generation, and orbital angular momentum (OAM) beam deflection at different angles of 0°, 10°, 15°, 25°, 30°, and 45°. A 3D-printed prototype of the dual-beam feature has been fabricated and measured for validation purposes, with good agreement between both simulation and measurement results, with small discrepancies due to 3D printing's low resolution and fabrication errors. This meta-lens shows promise for low-cost, high-gain beam deflection in mm-wave wireless communication systems, especially for sensing applications, with potential for wider 2D beam scanning and independent beam deflection enhancements.

Discovery of an Accretion Streamer and a Slow Wide-angle Outflow around FU Orionis

We present Atacama Large Millimeter/submillimeter Array 12-m, 7-m, and Total Power Array observations of the FU Orionis outbursting system, covering spatial scales ranging from 160 to 25,000 au. The high-resolution interferometric data reveal an elongated 12CO(2–1) feature previously observed at lower resolution in 12CO(3–2). Kinematic modeling indicates that this feature can be interpreted as an accretion streamer feeding the binary system. The mass infall rate provided by the streamer is significantly lower than the typical stellar accretion rates (even in quiescent states), suggesting that this streamer alone is not massive enough to sustain the enhanced accretion rates characteristic of the outbursting class prototype. The observed streamer may not be directly linked to the current outburst, but rather a remnant of a previous, more massive streamer that may have contributed enough to the disk mass to render it unstable and trigger the FU Orionis outburst. The new data detect, for the first time, a vast, slow-moving carbon monoxide molecular outflow emerging from this object. To accurately assess the outflow properties (mass, momentum, and kinetic energy), we employ 13CO(2–1) data to correct for optical depth effects. The analysis indicates that the outflow corresponds to swept-up material not associated with the current outburst, similar to the slow molecular outflows observed around other FUor and Class I protostellar objects.

Variable intraspecific response to climate change in a medicinally important African tree species, Vachellia sieberiana (DC.) (paperbark thorn).

Climate change is predicted to disproportionately impact sub-Saharan Africa, with potential devastating consequences on plant populations. Climate change may, however, impact intraspecific taxa differently. The aim of the study was to determine the current distribution and impact of climate change on three varieties of Vachellia sieberiana, that is, var. sieberiana, var. villosa and var. woodii. Ensemble species distribution models (SDMs) were built in "biomod2" using 66, 45, and 137 occurrence records for var. sieberiana, var. villosa, and var. woodii, respectively. The ensemble SDMs were projected to 2041-2060 and 2081-2100 under three general circulation models (GCMs) and two shared socioeconomic pathways (SSPs). The three GCMs were the Canadian Earth System Model version 5, the Institut Pierre-Simon Laplace Climate Model version 6A Low Resolution, and the Model for Interdisciplinary Research on Climate version 6. The suitable habitat of var. sieberiana predominantly occurs in the Sudanian and Zambezian phytochoria while that of var. villosa largely occurs in the Sudanian phytochorion. The suitable habitat of var. woodii mainly occurs in the Zambezian phyotochorion. There is coexistence of var. villosa and var. sieberiana in the Sudanian phytochorion while var. sieberiana and var. woodii coexist in the Zambezian phytochorion. Under SSP2-4.5 in 2041-2060 and averaged across the three GCMs, the suitable habitat expanded by 33.8% and 119.7% for var. sieberiana and var. villosa, respectively. In contrast, the suitable habitat of var. woodii contracted by -8.4%. Similar trends were observed in 2041-2060 under SSP5-8.5 [var. sieberiana (38.6%), var. villosa (139.0%), and var. woodii (-10.4%)], in 2081-2100 under SSP2-4.5 [var. sieberiana (4.6%), var. villosa (153.4%), and var. woodii (-14.4%)], and in 2081-2100 under SSP5-8.5 [var. sieberiana (49.3%), var. villosa (233.4%), and var. woodii (-30.7%)]. Different responses to climate change call for unique management and conservation decisions for the varieties.

Linking Mg ii and [O ii] spatial distribution to ionizing photon escape in confirmed LyC leakers and non-leakers

The geometry of the neutral gas in and around galaxies is a key regulator of the escape of ionizing photons. We present the first statistical study aimed at linking the neutral and ionized gas distributions to the Lyman continuum (LyC) escape fraction ( in a sample of 22 confirmed LyC leakers and non-leakers at $z using the Keck Cosmic Web Imager (Keck/KCWI) and the Low Resolution Spectrograph 2 (HET/LRS2). Our integral field unit data enable the detection of neutral and low-ionization gas, as traced by Mg ii and ionized gas, as traced by O ii extending beyond the stellar continuum for seven and ten objects, respectively. All but one object with extended Mg ii emission also show extended O ii emission; in this case Mg ii emission is always more extended than O ii by a factor 1.2 on average. Most of the galaxies with extended emission are non or weak LyC leakers ( < 5<!PCT!>), but we find a large diversity of neutral and low-ionization gas configurations around these weakly LyC-emitting galaxies. Conversely, the strongest leakers ( > 5<!PCT!>) appear uniformly compact in both Mg ii and O ii with exponential scale lengths lesssim 1 kpc. Most are unresolved at the resolution of our data. We also find a trend between and the spatial offsets of the nebular gas and the stellar continuum emission. Moreover, we find significant anticorrelations between the spatial extent of the neutral and/or low-ionization gas and the O iii O ii ratio, and Hbeta equivalent width, as well as positive correlations with metallicity and UV size, suggesting that galaxies with more compact neutral and/or low-ionization gas sizes are more highly ionized. The observations suggest that strong LyC emitters do not have extended neutral and/or low-ionization gas halos and ionizing photons may be emitted in many directions. Combined with high ionization diagnostics, we propose that the Mg ii and potentially O ii spatial compactness are indirect indicators of LyC emitting galaxies at high redshift.

Strategic optimization operations in the integrated energy system through multitime scale comprehensive demand response

AbstractIn the context of the dual carbon goal strategy, the proportion of new energy generation has increased annually, large‐scale renewable energy integration has been achieved, and the intermittent and uncertain operating characteristics pose an enormous challenge to the complete and stable operation of an integrated energy system (IES), promoting the complexity of IES optimization models. To increase the stability and accuracy of the system and improve the operation efficiency of the system, a Gaussian mixture model is used to fit the probability distribution of wind power and the prediction errors of the photovoltaic output. In addition, the expected maximization method is used to solve a model with hidden variables, the results show that the expectation maximization algorithm can improve the fitting accuracy, reduce the error caused by the subjective setting of the initial value, and make the fitting accuracy of the wind‐solar power prediction error greater. Then, to solve the problem in which the low resolution of the day‐ahead scheduling time leads to large errors, day‐ahead, day‐to‐day multitime scale operation optimization model takes into account the comprehensive demand response according to the differences in wind and solar output and load forecasting accuracy. Finally, simulation validation is conducted in multiple scenarios, and a comparative analysis is performed for single and multitime scale comprehensive demand response scenarios. The simulation results show that compared with the optimization results of no demand response day, the optimization results of this model reduce the total cost by 23.21%, carbon emissions by 13.98%, purchased electricity by 13.87%, and purchased gas by 19.31%, effectively improving the use of gas turbines in the system. The multitime scale scheduling strategy increases the overall operating cost of the IES, modifies the scheduling results, and agrees with real operating scenarios.

Research on Laser Dual-Mode Fusion Detection Method of Ship Wake Bubbles

Addressing the issues of weak echo signals and strong background interference in the laser detection of ships’ wakes, an analysis of the laser backscatter detection characteristics of ships’ wakes has been conducted. Based on the Monte Carlo method, a simulation model for the dual-mode fusion detection of ship wake bubbles using laser technology was constructed under different target characteristics. A dual-mode fusion detection system for ships’ wakes was designed, and an indoor experimental platform for the dual-mode fusion detection of ship wake bubbles using laser technology was established. To address problems such as a wide range of echo signal intensity changes, severe signal fluctuations, low resolution, poor image contrast, and blurred target edge information, an algorithm based on multi-timescale hierarchical fusion signal processing and temporal difference accumulation image processing was proposed. Verification experiments for ship wake detection were conducted, which revealed that the dual-mode fusion detection method for ship wake bubbles using laser technology can effectively enhance the detection signal-to-background ratio and counter the maneuvering evasion of underwater weapons by ships. It achieved high sensitivity, large dynamic range, high resolution, and a wide field of view detection and real-time signal processing of ship wake bubble targets of different magnitudes against a strong reverberation background. The effectiveness of the dual-mode fusion detection mode was validated, providing theoretical support for the overall system design and parameter settings.

Imaging 3D chemistry at 1 nm resolution with fused multi-modal electron tomography

Measuring the three-dimensional (3D) distribution of chemistry in nanoscale matter is a longstanding challenge for metrological science. The inelastic scattering events required for 3D chemical imaging are too rare, requiring high beam exposure that destroys the specimen before an experiment is completed. Even larger doses are required to achieve high resolution. Thus, chemical mapping in 3D has been unachievable except at lower resolution with the most radiation-hard materials. Here, high-resolution 3D chemical imaging is achieved near or below one-nanometer resolution in an Au-Fe3O4 metamaterial within an organic ligand matrix, Co3O4-Mn3O4 core-shell nanocrystals, and ZnS-Cu0.64S0.36 nanomaterial using fused multi-modal electron tomography. Multi-modal data fusion enables high-resolution chemical tomography often with 99% less dose by linking information encoded within both elastic (HAADF) and inelastic (EDX/EELS) signals. We thus demonstrate that sub-nanometer 3D resolution of chemistry is measurable for a broad class of geometrically and compositionally complex materials.

Selective activation of photoactivatable fluorescent protein based on binary holography

The ability to deliver laser doses to different target locations with high spatial and temporal resolution has been a long-sought goal in photo-stimulation and optogenetics research via, for example, photoactivatable proteins. These light-sensitive proteins undergo conformational changes upon photoactivation, serving functions such as triggering fluorescence, modulating ion channel activities, or initiating biochemical reactions within cells. Conventionally, photo-stimulation on light-sensitive proteins is performed by serially scanning a laser focus or via 2D projection, which is limited by relatively low spatiotemporal resolution. In this work, we present a programmable two-photon stimulation method based on a digital micromirror device (DMD) and binary holography to perform the activation of photoactivatable green fluorescent protein (PAGFP) in live cells. This method achieved grayscale and 3D selective PAGFP activation with subcellular resolution. In the experiments, we demonstrated the 3D activation capability and investigated the diffusion dynamics of activated PAGFP on the cell membrane. A regional difference in cell membrane diffusivity was observed, indicating the great potential of our approach in interrogating the spatiotemporal dynamics of cellular processes inside living cells.

3D Multiresolution Velocity Model Fusion with Probability Graphical Models

ABSTRACT The variability in spatial resolution of seismic velocity models obtained via tomographic methodologies is attributed to many factors, including inversion strategies, ray-path coverage, and data integrity. Integration of such models, with distinct resolutions, is crucial during the refinement of community models, thereby enhancing the precision of ground-motion simulations. Toward this goal, we introduce the probability graphical model (PGM), combining velocity models with heterogeneous resolutions and nonuniform data point distributions. The PGM integrates data relations across varying resolution subdomains, enhancing detail within low-resolution (LR) domains by utilizing information and prior knowledge from high-resolution (HR) subdomains through a maximum posterior problem. Assessment of efficacy, utilizing both 2D and 3D velocity models—consisting of synthetic checkerboard models and a fault-zone model from Ridgecrest, California—demonstrates noteworthy improvements in accuracy, compared to state-of-the-art fusion techniques. Specifically, we find reductions of 30% and 44% in computed travel-time residuals for 2D and 3D models, respectively, as compared to conventional smoothing techniques. Unlike conventional methods, the PGM’s adaptive weight selection facilitates preserving and learning details from complex, nonuniform HR models and applies the enhancements to the LR background domain.

Enhanced Video Super-Resolution Network towards Compressed Data

Video super-resolution (VSR) algorithms aim at recovering a temporally consistent high-resolution (HR) video from its corresponding low-resolution (LR) video sequence. Due to the limited bandwidth during video transmission, most available videos on the internet are compressed. Nevertheless, few existing algorithms consider the compression factor in practical applications. In this paper, we propose an enhanced VSR model towards compressed videos, termed as ECVSR, to simultaneously achieve compression artifacts reduction and SR reconstruction end-to-end. ECVSR contains a motion-excited temporal adaption network (METAN) and a multi-frame SR network (SRNet). The METAN takes decoded LR video frames as input and models inter-frame correlations via bidirectional deformable alignment and motion-excited temporal adaption, where temporal differences are calculated as motion prior to excite the motion-sensitive regions of temporal features. In SRNet, cascaded recurrent multi-scale blocks (RMSB) are employed to learn deep spatio-temporal representations from adapted multi-frame features. Then, we build a reconstruction module for spatio-temporal information integration and HR frame reconstruction, which is followed by a detail refinement module for texture and visual quality enhancement. Extensive experimental results on compressed videos demonstrate the superiority of our method for compressed VSR. Code will be available at https://github.com/lifengcs/ECVSR .

Investigation of 3D vessel reconstruction under Doppler imaging with phantoms: Towards reconstruction of the Circle of Willis

BackgroundStroke is the second leading cause of death across the globe. Early screening and risk detection could provide early intervention and possibly prevent its incidence. Imaging modalities, including 1D-Transcranial Doppler Ultrasound (1D-TCD) or Transcranial Color-code sonography (TCCS), could only provide low spatial resolution or 2D image information, respectively. Notably, 3D imaging modalities including CT have high radiation exposure, whereas MRI is expensive and cannot be adopted in patients with implanted devices. This study proposes an alternative imaging solution for reconstructing 3D Doppler ultrasound geared towards providing a screening tool for the 3D vessel structure of the brain. MethodsThe system comprises an ultrasound phased array attached to a servo motor, which can rotate 180˚ at a speed of 2˚/s. We extracted the color Doppler ROI from the image before reconstructing it into a 3D view using a customized pixel-based algorithm. Different vascular diameters, flow velocity, and depth were tested using a vascular phantom with a pumped flow to confirm the system for imaging blood flow. These variables were set to mimic the vessel diameter, flow speed, and depth of the Circle of Willis (CoW) during a transcranial screening. Results and conclusionsThe lower values of absolute error and ratio were found in the larger vascular channels, and vessel diameter overrepresentation was observed. Under different flow velocities, such diameter overrepresentation in the reconstructed flow did not change much; however, it did change with different depths. Meanwhile, the setting of the velocity scale and the color gain affected the dimension of reconstructed objectives. Moreover, we presented a 3D image of CoW from a subject to demonstrate its potential. The findings of this work can provide a good reference for further studies on the reconstruction of the CoW or other blood vessels using Doppler imaging.

WindSeer: real-time volumetric wind prediction over complex terrain aboard a small uncrewed aerial vehicle

Real-time high-resolution wind predictions are beneficial for various applications including safe crewed and uncrewed aviation. Current weather models require too much compute and lack the necessary predictive capabilities as they are valid only at the scale of multiple kilometers and hours – much lower spatial and temporal resolutions than these applications require. Our work demonstrates the ability to predict low-altitude time-averaged wind fields in real time on limited-compute devices, from only sparse measurement data. We train a deep neural network-based model, WindSeer, using only synthetic data from computational fluid dynamics simulations and show that it can successfully predict real wind fields over terrain with known topography from just a few noisy and spatially clustered wind measurements. WindSeer can generate accurate predictions at different resolutions and domain sizes on previously unseen topography without retraining. We demonstrate that the model successfully predicts historical wind data collected by weather stations and wind measured by drones during flight.

Mercury and selenium distribution in human brain tissue using synchrotron micro-X-ray fluorescence.

Mercury is a well-recognized environmental contaminant and neurotoxin, having been associated with a number of deleterious neurological conditions including neurodegenerative diseases, such as Alzheimer's disease. To investigate how mercury and other metals behave in the brain, we used synchrotron micro-X-ray fluorescence to map the distribution pattern and quantify concentrations of metals in human brain. Brain tissue was provided by the Rush Alzheimer's Disease Center and samples originated from individuals diagnosed with Alzheimer's disease and without cognitive impairment. Data were collected at the 2-ID-E beamline at the Advanced Photon Source at Argonne National Laboratory with an incident beam energy of 13keV. Course scans were performed at low resolution to determine gross tissue features, after which smaller regions were selected to image at higher resolution. The findings revealed (1) the existence of mercury particles in the brain samples of two subjects; (2) co-localization and linear correlation of mercury and selenium in all particles; (3) co-localization of these particles with zinc structures; and (4) association with sulfur in some of these particles. These results suggest that selenium and sulfur may play protective roles against mercury in the brain, potentially binding with the metal to reduce the induced toxicity, although at different affinities. Our findings call for further studies to investigate the relationship between mercury, selenium, and sulfur, as well as the potential implications in Alzheimer's disease and related dementias.

HR-YOLOv8: A Crop Growth Status Object Detection Method Based on YOLOv8

Crop growth status detection is significant in agriculture and is vital in planting planning, crop yield, and reducing the consumption of fertilizers and workforce. However, little attention has been paid to detecting the growth status of each crop. Accuracy remains a challenging problem due to the small size of individual targets in the image. This paper proposes an object detection model, HR-YOLOv8, where HR means High-Resolution, based on a self-attention mechanism to alleviate the above problem. First, we add a new dual self-attention mechanism to the backbone network of YOLOv8 to improve the model’s attention to small targets. Second, we use InnerShape(IS)-IoU as the bounding box regression loss, computed by focusing on the shape and size of the bounding box itself. Finally, we modify the feature fusion part by connecting the convolution streams from high resolution to low resolution in parallel instead of in series. As a result, our method can maintain a high resolution in the feature fusion part rather than recovering high resolution from low resolution, and the learned representation is more spatially accurate. Repeated multiresolution fusion improves the high-resolution representation with the help of the low-resolution representation. Our proposed HR-YOLOv8 model improves the detection performance on crop growth states. The experimental results show that on the oilpalmuav dataset and strawberry ripeness dataset, our model has fewer parameters compared to the baseline model, and the average detection accuracy is 5.2% and 0.6% higher than the baseline model, respectively. Our model’s overall performance is much better than other mainstream models. The proposed method effectively improves the ability to detect small objects.

A flow-based multi-scale learning network for single image stochastic super-resolution

Single image super-resolution (SISR) is still an important while challenging task. Existing methods usually ignore the diversity of generated Super-Resolution (SR) images. The fine details of the corresponding high-resolution (HR) images cannot be confidently recovered due to the degradation of detail in low-resolution (LR) images. To address the above issue, this paper presents a flow-based multi-scale learning network (FMLnet) to explore the diverse mapping spaces for SR. First, we propose a multi-scale learning block (MLB) to extract the underlying features of the LR image. Second, the introduced pixel-wise multi-head attention allows our model to map multiple representation subspaces simultaneously. Third, by employing a normalizing flow module for a given LR input, our approach generates various stochastic SR outputs with high visual quality. The trade-off between fidelity and perceptual quality can be controlled. Finally, the experimental results on five datasets demonstrate that the proposed network outperforms the existing methods in terms of diversity, and achieves competitive PSNR/SSIM results. Code is available at https://github.com/qianyuwu/FMLnet.