Search

Individual tree detection and counting in unmanned aerial vehicle (UAV) imagery constitute a vital and practical research field. Vegetation remote sensing captures large-scale trees characterized by complex textures, significant growth variations, and high species similarity within the vegetation, which presents significant challenges for annotation and detection. Existing methods based on bounding boxes have struggled to convey semantics information about tree crowns. This paper proposes a novel deep learning network called VrsNet based on the density map information. The proposed work pioneers the segmentation and counting application by utilizing the semantic information of Gaussian contour. Besides, we sample and create the UAV vegetation remote sensing density dataset TreeFsc for experiments. In quantitative comparison across multiple datasets, the proposed method demonstrates high performance, with a 3.45 increase in MAE and a 4.75 increase in RMSE. Experiments demonstrate superior cross-region, cross-scale, and cross-species target detection capabilities of the proposed approach compared with the existing object detection methods. Our code and dataset are available at: https://github.com/luotiger123/VrsNet/tree/main/VrsNet.

Image detection technology is of paramount importance across various fields. This significance is not only seen in general images with everyday scenes but also holds substantial research value in the field of remote sensing. Remote sensing images involve capturing images from aircraft or satellites. These images typically feature diverse scenes, large image formats, and varying imaging heights, thus leading to numerous small-sized targets in the captured images. Accurately identifying these small targets, which may occupy only a few pixels, is a challenging and active research area. Current methods mainly fall into two categories: enhancing small target features by improving resolution and increasing the number of small targets to bolster training datasets. However, these approaches often fail to address the core distinguishing features of small targets in the original images, thus resulting in suboptimal performance in fine-grained classification tasks. To address this situation, we propose a new network structure DDU (Downsample Difference Upsample), which is based on differential and resolution changing methods in the Neck layer of deep learning networks to enhance the recognition features of small targets, thus further improving the feature richness of recognition and effectively solving the problem of low accuracy in small target object recognition. At the same time, in order to take into account the recognition effect of targets of other sizes in the image, a new attention mechanism called PNOC (protecting the number of channels) is proposed, which integrates small target features and universal object features without losing the number of channels, thereby increasing the accuracy of recognition. And experimental verification was conducted on the PASCAL-VOC dataset. At the same time, it was applied to the testing of the fine-grained MAR20 dataset and found that the performance was better than other classic algorithms. At the same time, because the proposed framework belongs to a one-stage detection method, it has good engineering applicability and scalability, and universality in scientific research applications are good. Through comparative experiments, it was found that our algorithm improved the performance of the mAP by 0.7% compared to the original YOLOv8 algorithm.

Owing to the structural errors in the optical phased array, an initial random phase reduces the quality of the deflection beam. The most commonly applied approach to phase calibration is based on adaptive optics. However, adaptive optimisation approaches have slow convergence and low diffraction efficiency. We proposed a pointwise optimisation approach to achieve fast and accurate beam deflection. This approach conducts phase calibration, combining global traversal and local searches individually for each array element. We built a phase-calibration optical system containing a one-dimensional optical waveguide phase array for further verification and designed the relevant mechanics. The simulation and experimental results demonstrate that the pointwise optimisation approach accelerates the calibration process and improves the diffraction efficiency.

CIOMP report their progress in high-accuracy manufacturing of the world’s largest SiC aspheric mirror. The final surface figure error is 15.2 nm RMS and aerial density is less than 120 kg/m2.

The application of porous carbon microspheres derived from pure biomass in supercapacitors is restricted due to their limited reactive groups. MXene owns a combination of redox Faradic surface with good metallic conductivity and hydrophilicity, which assists to obtain high pseudocapacitance and energy density. Herein, Ti3C2Tx MXene was introduced to chitosan-based porous carbon microsphere (CPCM) to fabricated sandwich-like structure (CPCM/MXene) through electrostatic interaction. The Ti3C2Tx protected the spherical structure of CPCM. Meanwhile, CPCM hindered the reaggregation of Ti3C2Tx by inserting in the Ti3C2Tx layers, promoting the electrolyte migration kinetics. The synergistic effect endowed CPCM/MXene high specific capacitance of 362 F/g at current density of 0.5 A/g and acceptable cycling stability with 93.87% capacitance retention at a high current density of 10 A/g after 10,000 cycles. Furthermore, CPCM/MXene displayed a high energy density of 27.8 W/(h•kg) at 500.0 W/kg of power density. These satisfactory performances prove that combining Ti3C2Tx MXene nanosheets with porous carbon microspheres is a considering method to construct a new generation electrode material of supercapacitor.

To determine furaltadone metabolites 5-methylmorpholino-3-amino-2-oxazolidinone (AMOZ) residual levels in ctenopharyngodon idellus, direct competitive enzyme-linked immunosorbent assay (dcELISA) was developed based on a single-chain variable fragment (scFv)-alkaline phosphatase (AP) fusion protein. Firstly, VH and VL gene sequences were cloned from hybridoma cell lines secreting monoclonal antibody against AMOZ derivative, and then the productive VH and VL genes were assembled to a scFv gene and engineered to produce scFv-AP fusion protein. The fusion protein was further identified as a bifunctional reagent for immunoassay, then a sensitive one-step dcELISA against AMOZ derivative was developed with a half-maximal inhibitory concentration (IC50) and limit of detection (LOD) of 10.62 and 4.83 ng/mL, respectively. The recovery values of AMOZ from spiked fish samples determined by dcELISA were in 71.31%∼82.88%. The results suggested that the prepared scFv-AP fusion protein in this work can be used for rapid and convenient immunoassays to detect AMOZ residues in fish samples.

Singlet microscopy is very attractive for the development of cost-effective and portable microscopes. In contrast to conventional microscope objectives, which consist of multiple lenses, the manufacturing process for singlet lenses is done without extensive assembling and aligning. In this manuscript, we report a novel singlet virtual Zernike phase contrast microscopy setup for unstained pathological tumor tissue slides. In this setup, the objective consists of only one lens. There is no need for the inset Zernike phase plate, which is even more expensive than a whole brightfield microscopy setup. The Zernike phase contrast is virtually achieved by the deep learning computational imaging method. For the practical virtual Zernike phase contrast microscopy setup, the computational time is less than 100 ms, which is far less than that of other computational quantitative phase imaging algorithms. With a conceptual demo experimental setup, we proved our proposed method to be competitive with a research-level conventional Zernike phase contrast microscope and effective for the unstained transparent pathological tumor tissue slides. It is believed that our deep learning singlet virtual phase contrast microscopy is potential for the development of low-cost and portable microscopes and benefits resource-limited areas.

Three-dimensional printing, also called additive manufacturing (AM), offers a new vision for optical components in terms of weight reduction and strength improvement. A truss, which is a triangulated system of members that are structured and connected in such a way that they mainly bear axial force, is commonly used in steel structures to improve stiffness and reduce weight. Combining these two technologies, an extremely lightweight truss-structured mirror was proposed. First, the finite element analyses (FEA) on surface shape deviation and modal properties were carried out. Results showed that the mirrors had sufficient stiffness and a high weight reduction of up to 85%. In order to verify their performance, the truss-structured mirror blanks were fabricated with AM technology. After that, both the preprocessing and the postprocessing of the mirrors were carried out. The results show that without NiP coating, a surface shape deviation of 0.353λ (PV) and 0.028 λ (RMS) (λ = 632.8 nm) with a roughness of Ra 2.8 nm, could be achieved. Therefore, the truss-structured mirrors in this study have the characteristics of being extremely lightweight and having improved stiffness as well as strong temperature stability.

Singlet lenses are free from precise assembling, aligning, and testing, which are helpful for the development of portable and low-cost microscopes. However, balancing the spectrum dispersion or chromatic aberrations using a singlet lens made of one material is difficult. Here, a novel method combining singlet lens microscopy and computational imaging, which is based on deep learning image-style-transfer algorithms, is proposed to overcome this problem in clinical pathological slide microscopy. In this manuscript, a singlet aspheric lens is used, which has a high cut-off frequency and linear signal properties. Enhanced by a trained deep learning network, it is easy to transfer the monochromatic gray-scale microscopy picture to a colorful microscopy picture, with only one single-shot recording by a monochromatic CMOS image sensor. By experiments, data analysis, and discussions, it is proved that our proposed virtual colorization microscope imaging method is effective for H&E stained tumor tissue slides in singlet microscopy. It is believable that the computational virtual colorization method for singlet microscopes would promote the low-cost and portable singlet microscopy development in medical pathological label staining observing (e.g., H&E staining, Gram staining, and fluorescent labeling) biomedical research.

In computer-controlled optical surfacing (CCOS), the paths of the lap tools are limited inside the optical surface; this restricts convolution in the dwell-time algorithm and causes mid-spatial-frequency surface errors. An optimized strategy ensuring relatively complete convolution of the dwell-time algorithm is developed to control the mid-spatial-frequency surface error and simultaneously ensure high optical manufacturing efficiency. Different-sized lap tools are then introduced to correct the surface error in different areas of the optics. Simulations and experiments using a large off-axis SiC mirror demonstrate the validity of the strategy, and it could be widely applied to CCOS in grinding or polishing processes.

Magnetorheological finishing (MRF) fabrication can effectively facilitate a rapid fabrication of an aspherical 2 m reaction-bonded silicon carbide mirror. The dwell-time algorithm and tool path supported tool-mark mitigation and virtual-axis employment are analyzed. A rapid-fabrication strategy alternates MRF and a large polishing lap to converge surface error profile before MRF alone is used to high-precise finishing. In the alternate-use process, large polishing lap’s dwell time map for low-order space frequency is calculated before MRF’s for high-order one, but executed later in reality. The fabricated mirror with a silicon modification layer showed accuracy convergence from 0.098 λ rms to 0.019 λ rms at 84.6 h, demonstrating the strategy’s validity for large optical-surface processing. Keywords: Computer controlled polishing, Silicon carbide mirror, Surface error, Magnetorheological finishing, Large polishing lap

Er ^3+ /Yb ^3+ co-doped TiO _2 –ZrO _2 –La _2 O _3 glasses were prepared by aerodynamic levitation method. The transmittance spectrum shows that the transmittance of the glass before and after irradiation decreases from 75.46% to 71.13% in visible and near-infrared bands. The absorption and emission cross sections of the glasses at 525 nm were studied. The glasses were thermally treated at various temperatures and then investigated by x-ray diffraction, Scanning Electron Microscope and Transmission Electron Microscope. Optical spectra were measured as a function of temperature in the range of 298–498 K. The maximum relative sensitivity was determined to be 1.47% K ^−1 at 298 K. These results imply that the obtained glass may have the potential to be the matrix material for temperature sensing that requires to employed in a complex environment.