Search

Reducing particle size in supported metal catalysts to single-atom level isolates the active metal sites and maximizes the atomic utilization efficiency. However, the large inter-atom distance, particularly in low-loading single-atom catalyst (SAC), is not favorable for a complex reaction where two (or more) reactants have to be activated. A key question is how to control the inter-atom distances to promote dinuclear-type coactivation at the adjacent metal sites. Here, it is reported that reducing the average inter-atom distance of copper SACs supported on carbon nitride (C

Comparatively revealing the complexation behavior of trivalent actinides and lanthanides with functional ligands in aqueous solution is of great importance to enrich our knowledge on the fundamental coordination chemistry of trivalent f-block elements and to control the fate of minor actinides in nuclear fuel cycles. In this work, the complexation of Am(III) and Nd(III), representatives for trivalent actinides and lanthanides, respectively, with a N,O-hybrid ligand 6-(dimethylcarbamoyl)picolinic acid (DMAPA, denoted as HL) was investigated by absorption spectroscopy, calorimetry, X-ray crystallography, and density functional theory (DFT) calculations. Successive formation of 1:1, 1:2, and 1:3 (metal/ligand) complexes of Am(III) and Nd(III) with DMAPA was identified, and the corresponding thermodynamic parameters were determined. The binding strength of Am(III) with DMAPA is slightly stronger than that of Nd(III), and the complexation of Nd(III) with DMAPA is mainly entropy-driven. The crystal structure of the 1:2 Nd(III)/DMAPA complex and the DFT calculation shed additional light on the coordination and structural characteristics of the complexes. In contrast to the Nd-N bond in the Nd(III)/DMAPA complex, the Am-N bond in the Am(III)/DMAPA complex exhibits more covalency, which contributes to the slightly stronger complexation of Am(III) with DMAPA.

For pre-operative osteoarthritis (OA) patients with varus knee, previous studies showed inconsistent results. Therefore, we conducted this study to better identify the association of Hospital for Special Surgery (HSS) score and mechanical alignment. 44 patients (51 knees) with constitutional varus knee caused by combined deformities (LDFA (lateral distal femoral angle) > 90°and MPTA (medial proximal tibial angle) p = 0.5370). For pre-operative constitutional varus knee caused by combined deformities in chinese populations, no significant association between post-operative lower limb mechanical alignment and HSS score was found.

Diabetic retinopathy is one of the main causes of blindness in human eyes, and lesion segmentation is an important basic work for the diagnosis of diabetic retinopathy. Due to the small lesion areas scattered in fundus images, it is laborious to segment the lesion of diabetic retinopathy effectively with the existing U-Net model. In this paper, we proposed a new lesion segmentation model named FFU-Net (Feature Fusion U-Net) that enhances U-Net from the following points. Firstly, the pooling layer in the network is replaced with a convolutional layer to reduce spatial loss of the fundus image. Then, we integrate multiscale feature fusion (MSFF) block into the encoders which helps the network to learn multiscale features efficiently and enrich the information carried with skip connection and lower-resolution decoder by fusing contextual channel attention (CCA) models. Finally, in order to solve the problems of data imbalance and misclassification, we present a Balanced Focal Loss function. In the experiments on benchmark dataset IDRID, we make an ablation study to verify the effectiveness of each component and compare FFU-Net against several state-of-the-art models. In comparison with baseline U-Net, FFU-Net improves the segmentation performance by 11.97%, 10.68%, and 5.79% on metrics SEN, IOU, and DICE, respectively. The quantitative and qualitative results demonstrate the superiority of our FFU-Net in the task of lesion segmentation of diabetic retinopathy.

Super resolution (SR) and Perceptual Image Enhancement (PIE) are gaining more and more interests in digital image processing and have been studied independently in the past decades. Although plenty of state-of-the-art researches have demonstrated great improvement in SR problem, they neglect practical requirements in real-world application. In practice, these two tasks are always mixed and combined to obtain a high-resolution enhanced (HRE) image with high quality from a low-resolution original image (LRO) with low quality. In this paper, we propose a joint SR-PIE learning framework called Deep SR-PIE, which comprises Multi-scale Backward Fusion Network (MBFNet), Perceptual Enhancement Network (PENet) and Dual-Path Unsampling Network (DUNet). MBFNet network is responsible for deep feature representation for further image reconstruction and perceptual enhancement, and PENet seeks the optimal local transformation to recover perceptual loss (color, tone, exposure and so on). DUNet works in different scales and exchanges each other to complement more details during upsampling. In our experiments, a real-world dataset is released to facilitate the development of joint learning for SR and PIE. Then, a thorough ablation study is provided to better understand the superiority of our method. Finally, extensive experiments suggest that the proposed method performs favorably against the state-of-the-arts in terms of visual quality, PSRN, SSIM, model size and inference time. By virtue of splitting operation and inverse residual blocks, as a lightweight deep neural network, our model is compatible with low-computation device.

Technology has made life simpler for every human being, irrespective of educational qualifications and sector of work. The education sector in the current modern era is revolutionized by technology. Teaching and understanding theoretical subjects like ideological and pedagogical education (IaPE) has become simple with technologies like artificial intelligence (AI), edge computing (EC), and flipped classrooms. These are unique technologies in the education sector to make classroom learning more manageable by creating concepts in digitization and visualization with audiovisual technologies. In recent research, many authors have proposed models for IaPE through ideological and pedagogical teaching (IaPT) for different courses in vocational colleges and universities. Some researchers have designed a hybrid model with IaPT and analyzed the results as an extension to this teaching method. This research work introduced AI and EC technologies to the flipped classroom to promote higher vocational education in colleges and universities for IaPE. With digitization and visualization in course material preparation, AI aids in an interactive classroom study. On the other hand, EC aids in the easy storage and accessing of the resources prepared through AI. The analysis work is performed on the dataset from the unique client identifier (UCI) repository by proposing a novel optimization algorithm for flipped classrooms.

The knee joint, as the main lower limb motor joint, is the most vulnerable and susceptible joint. The knee injuries considerably impact the normal living ability and mental health of patients. Understanding the biomechanics of a normal and diseased knee joint is in urgent need for designing knee assistive devices and optimizing a rehabilitation exercise program. In this paper, we systematically searched electronic databases (from 2000 to November 2019) including ScienceDirect, Web of Science, PubMed, Google Scholar, and IEEE/IET Electronic Library for potentially relevant articles. After duplicates were removed and inclusion criteria applied to the titles, abstracts, and full text, 138 articles remained for review. The selected articles were divided into two groups to be analyzed. Firstly, the real movement of a normal knee joint and the normal knee biomechanics of four kinds of daily motions in the sagittal and coronal planes, which include normal walking, running, stair climbing, and sit-to-stand, were discussed and analyzed. Secondly, an overview of the current knowledge on the movement biomechanical effects of common knee musculoskeletal disorders and knee neurological disorders were provided. Finally, a discussion of the existing problems in the current studies and some recommendation for future research were presented. In general, this review reveals that there is no clear assessment about the biomechanics of normal and diseased knee joints at the current state of the art. The biomechanics properties could be significantly affected by knee musculoskeletal or neurological disorders. Deeper understanding of the biomechanics of the normal and diseased knee joint will still be an urgent need in the future.

Uncovering the complexation properties of actinides with common organic ligands in aqueous solution is of significant importance to understand the fundamental coordination chemistry of actinides as well as to develop more efficient extractants for actinides separation in advanced nuclear fuel cycles. Herein, the complexation behaviour of hexavalent plutonyl (Pu(VI) or PuO22+) with three dicarboxylic acids, iminodiacetic acid (IDA), N-methyl-iminodiacetic acid (MIDA) and dipicolinic acid (DPA), was studied through both experimental and theoretical approaches. In 1.0 M NaClO4 medium, successive formation of 1 : 1 and 1 : 2 (metal/ligand ratio) complexes of Pu(VI) with IDA and MIDA were identified and their stability constants were determined and compared with those of DPA. The complexation strength varies for the complexes of Pu(VI) with IDA, MIDA and DPA. For the 1 : 1 complexes, the stability constants of Pu(VI) with the three ligands follow the order of IDA < MIDA < DPA, while those of the 1 : 2 complexes follow the order of IDA < DPA < MIDA. The complexation trends are validated by DFT calculations, which suggest that the complexation strength of IDA/MIDA/DPA with Pu(VI) is governed by both the electronic effect (electron withdrawn group or electron donating group) and the rigidity of the ligands. Furthermore, the symmetric features of the Pu(VI) complexes with the three ligands were illustrated by analyzing the optical characteristics of the complexes and DFT calculation. The results provide new insights into the intrinsic mechanism of hexavalent actinide complexation with tridentate dicarboxylic acids in aqueous solutions.

As a detection tool to identify metal or alloy, metallographic quantitative analysis has received increasing attention for its ability to evaluate quality control and reveal mechanical properties. The detection procedure is mainly operated manually to locate and characterize the constitution in metallographic images. The automatic detection is still a challenge even with the emergence of several excellent models. Benefiting from the development of deep learning, with regard to two different metallurgical structural steel image datasets, we propose two attention-aware deep neural networks, Modified Attention U-Net (MAUNet) and Self-adaptive Attention-aware Soft Anchor-Point Detector (SASAPD), to identify structures and evaluate their performance. Specifically, in the case of analyzing single-phase metallographic image, MAUNet investigates the difference between low-frequency and high-frequency and prevents duplication of low-resolution information in skip connection used in an U-Net like structure, and incorporates spatial-channel attention module with the decoder to enhance interpretability of features. In the case of analyzing multi-phase metallographic image, SASAPD explores and ranks the importance of anchor points, forming soft-weighted samples in subsequent loss design, and self-adaptively evaluates the contributions of attention-aware pyramid features to assist in detecting elements in different sizes. Extensive experiments on the above two datasets demonstrate the superiority and effectiveness of our two deep neural networks compared to state-of-the-art models on different metrics.

Photoperiodism is a vital environmental factor that regulates the growth, development, metabolism, and reproduction of aquatic organisms. Here, we examined how different photoperiods (0L: 24D, 8L: 16D, 12L: 12D, 16L: 8D, 24L: 0D) with a full spectrum LED, influence growth, feeding, digestive enzymes, and antioxidant enzymes in juvenile European Sea Bass (Dicentrachus labrax L.). Specifically, a total of 300 fishes (body length: 13.5 ± 0.54 cm, bodyweight: 45 ± 5.3 g) were randomly divided into 5 different light treatment groups. Sampling and analysis were conducted after 60 days of the experiment. Results revealed significantly higher weight gain, specific growth rate, and daily growth index in the 8L: 16D group, relative to 0L: 24D and 24L: 0D. Conversely, feed conversion ratio (FCR) was significantly lower in 8L: 16D, 12L: 12D, and 16L: 8D treatments than in 24L: 0D and 0L: 24D. Additionally, activities of intestinal amylase (AMS) lipase (LPS) pepsin (PPS) and trypsin (TPS) in the stomach were significantly higher in the 8L: 16D, relative to the other treatment groups. Gills of juveniles exposed to 0L: 24D recorded significantly higher catalase (CAT) and superoxide dismutase (SOD) activities, compared to those in the other groups. Regarding the dorsal muscle, a significantly higher CAT activity was recorded under 0L: 24D light than the other light photoperiods. Moreover, kidneys of juveniles exposed to 24L: 0D recorded the highest CAT and SOD activity. In summary, 8L: 16D, 12L: 12D, and 16L: 8D light photoperiods increased the digestive enzyme activity, reduced oxidative stress, and enhanced growth in fish, despite long-term exposure to light or darkness being unsuitable. These findings indicate that light and dark combinations are more conducive for culturing sea bass, with the 8L: 16D regime being the optimal photoperiod for the growth of juveniles.

Kandelia obovata (Ko) and Aegiceras corniculatum (Ac) are common and dominant plant species in mangrove wetlands in South China, which are distributed in similar tidal zones along the coastline. The present study aimed to determine the allelopathic effects of leaf litter leachates (LLLs) from Ko and their purified condensed tannins (PCTs) on the germination and growth of Ac by mangrove microcosms. Replicate pots containing five different levels of LLLs and PCTs were separately prepared and propagules of Ac were placed in each treatment. Both LLLs and PCTs significantly inhibited the germination and growth of Ac, especially at high concentrations. The final germination rates of the roots and stems and the numbers of fine roots declined continuously, while other growth indicators, including the lengths of fine roots and nutritive roots and the biomasses of roots, stems, and leaves first increased and then decreased with increasing levels. These results indicate that LLLs from the leaf litter of Ko, in particular their PCTs, exerted an inhibition effect on propagule germination and seedling growth of Ac, and the inhibitory effects were concentration dependent. This study suggested that condensed tannins from leaf litter, acting as allelochemicals, could regulate the natural regeneration of a mangrove forest.

The analysis of cancer-associated genetic copy number variations (CNVs) has been employed for cancer diagnostics, treatment, and prognostic assessments. The use of liquid biopsies to identify mutated gene biomarkers is of clinical significance to cancer diagnostics, but the methodology of the procedure remains challenging. In this study, we establish ligation-mediated droplet digital PCR for the highly sensitive and precise quantification of CNVs which enables precise copy number assessments in various biosamples at the single-molecule level. In virtue of its excellent performances, circulating-free target DNA in cell or plasma samples can be assessed, permitting a more detailed analysis of cancer-related genetic alterations. The proposed assay provides a useful tool to reveal how CNVs influence the occurrence of cancer in liquid biopsy samples, which is of clinical significance to cancer diagnostics and treatment.

At present, there are some key issues in the traditional preimmersion method for reducing the collapsibility of the loess ground, such as the difficulty in determining the total water consumption and the long immersion time. In response to these issues, a new method, the borehole preimmersion method, is presented, and a specific theoretical design model is proposed for application in projects. The method is specifically discussed from a new perspective, and the diffusion mechanism and evolution law of water in the ground are presented in detail through theoretical analysis and numerical calculation, respectively. The water diffusion is a mushroom-type form for a single water injection hole immersed in water. A calculation model derived for a single water injection hole or a group of water injection holes based on the research results is used to calculate the volumes of soaked loess and the total water consumption. Through an in situ immersion test, the treatment effect of this method is evaluated to verify the rationality of the method and the theoretical calculation model proposed in this study, which provides a new method and theoretical framework for effectively reducing the collapsibility of the loess ground.

In the geogenetic overburden excavation of underground space projects (such as comprehensive pipe corridors or subways), the foundation stress is in an unloading state. The effect(s) from the unloading on the coefficient of subsidence in the loess area should be considered. In this study, to explore the effect of unloading on the collapsibility of loess, the collapsible loess in the Guanzhong area was considered as the research object. An expression for the unloading collapse coefficient was established based on the unloading stress ratio, unloading collapse ratio, and other parameters. The influence of the unloading on the loess collapse coefficient was studied using an indoor collapsibility test, and the function form and parameters for the expression were determined. As combined with the field test, the accuracy of the expression for the unloading coefficient was verified based on the test value for the specific collapsibility, calculated value for the specific collapsibility, and calculated value for the unloading collapsibility.

In this study, the impacts of various illumination intensities on growth, digestion and metabolism as well as antioxidant enzyme dynamics of Takifugu rubripes (0.03 ± 0.01 g) were investigated. Activities of alkaline phosphatase (AKP), acid phosphatase (ACP), pepsin (PPS), alanine aminotransferase (ALT), aspartate aminotransferase (AST), lactate dehydrogenase (LDH), superoxide dismutase (SOD), and total antioxidant capacity (T-AOC) were measured. Juveniles were reared under four degrees of illumination intensities: 50 mW/m2, 250 mW/m2, 500 mW/m2 and 750 mW/m2, at a constant photoperiod (12D:12 L). Samples were collected at days 7, 14, 21, 28, and 35. After 35 days, the 750 mW/m2 light intensity was found to have significantly increased the final body weight, which was 1.1 ± 0.005 g. On day 35, PPS activities of T. rubripes juveniles in the 750 mW/m2 sample group were significantly low when compared to those of the 250 mW/m2 group while AKP activities in the 750 mW/m2 group were significantly higher than those of the 50 and 500 mW/m2 groups (p 0.05). Even though they were relatively lower, there were no significant differences in SOD activities in the 250 mW/m2 group relative to the 500 and 750 mW/m2 groups. On day 35, high light intensity (750 mW/m2) resulted in stress responses that were characterized by T-AOC activities, and activity values were significantly lower than those of other groups (p

Aims Non-small cell lung cancer (NSCLC) is considered a highly fatal tumor. Importantly, angiogenesis is critical for tumor progression. Long non-coding RNAs (lncRNAs), which are untranslatable, control cell functions through different pathways. lncRNA EPIC1 has been reported to promote cell viability, cell cycle progression, and invasion. However, the relationship between EPIC1 and tumor angiogenesis remains an enigma. We explored the role of EPIC1 in tumor angiogenesis in NSCLC. Materials and methods First, EPIC1 expression was analyzed using the GEPIA database and was further verified using qPCR in tumor tissues from patients with NSCLC and NSCLC cell lines. Next, EPIC1 function was detected using loss-of-function and gain-of-function assays. Moreover, EdU staining, flow cytometry, and channel formation assays were performed to assess HUVEC proliferation and channel the formation in the NSCLC-HUVEC transwell co-culture system. Key findings EPIC1 expression was significantly upregulated in NSCLC tissues and cell lines. Furthermore, the overexpression of EPIC1 in NSCLC cells stimulated HUVEC channel formation and proliferation by activating Ang2/Tie2 signaling, and the opposite results were obtained when EPIC1 was silenced in NSCLC cells. The density of new blood vessels was simultaneously increased by EPIC1 overexpression in vivo, using CAM angiogenesis model and a nude mouse tumor model. Finally, all these experimental findings could be established in the samples from patients with NSCLC. We postulate that EPIC1 promotes tumor angiogenesis by activating the Ang2/Tie2 axis in NSCLC. Significance Elucidating the molecular and cellular mechanisms of EPIC1 in tumor angiogenesis provides a novel perspective on NSCLC clinical therapy.

Mangrove has been destroyed and reforestation is often undertaken, but whether a regenerated forest could restore its ecological function is not clear. This study compares microbial community structure and function in sediment of the 17-years old natural regenerated mangrove forest (Y17) with the original forest (Y74). No significant differences in phospholipid fatty acid (PLFA) profiles and microbial metabolism of most carbon substrates were found between these two forests. However, activities of dehydrogenase, protease, cellulase and phosphatase were lower in Y17 than Y74, and some specific microbial functions were also different. Both forests exhibited significant seasonal differences in enzyme activities and microbial characteristics, but such difference was larger in Y17 than Y74, indicating the regenerated forest was more sensitive to season. Correspondence analysis based on PLFA profiles and enzyme activities revealed the microbial community in Y17 was comparable to Y74, suggesting sediment microbial characteristics in natural regenerated mangroves could be restored.

CT images are commonly used in medical clinical diagnosis. However, due to factors such as hardware and scanning time, CT images in real scenes are limited by spatial resolution so that doctors cannot perform accurate disease analysis on tiny lesion areas and pathological features. An image super-resolution (SR) method based on deep learning is a good way to solve this problem. Although many excellent networks have been proposed, but they all pay more attention to image quality indicators than image visual perception quality. Unlike other networks that focus more on image evaluation metrics, the super resolution generative adversarial network (SRGAN) has achieved tremendous improvements in image perception quality. Based on the above, this paper proposes a CT image super-resolution algorithm based on improved SRGAN. In order to improve the visual quality of CT images, a dilated convolution module is introduced. At the same time, in order to improve the overall visual effect of the image, the mean structural similarity (MSSIM) loss is also introduced to improve the perceptual loss function. Experimental results on the public CT image dataset demonstrate that our model is better than the baseline method SRGAN not only in mean opinion score(MOS), but also in peak signal-to-noise ratio (PSNR) and structural similarity (SSIM) values.

The selective separation of trivalent americium from lanthanides in a nitric acid medium by a tetradentate ligand, N,N′-diethyl-N,N′-ditolyl-2,9-diamide-1,10-phenanthroline (Et-Tol-DAPhen), in an i...

The loss of coastal wetlands in Hong Kong Mai Po Nature Reserve adversely affected wetland-depended species. To mitigate this impact, gei wai ponds were reconstructed according to a set of biodiversity management zones (BMZs). This study, based on Canonical Correspondence Analysis (CCA), investigated if waterbird distribution was related to BMZ characteristics. Based on habitat characteristics, ponds in the same BMZ generally clumped in the same quadrant or within a short distance on CCA scatter plots, indicating that a BMZ zone produced common habitat traits. Ponds in a close distance on the plot had similar bird abundance or community structure. Significant correlations were noted between the abundance of cormorants and tall tree, and between waders and bare ground areas within study ponds. This study indicated that the control of key habitat factors was important for the success of reconstruction of gei wais and management of waterbirds in Mai Po.

Photoperiod is one of the important environmental factors, which have important effects on the early growth and development, survival, metabolism and immune system of larvae. In the present study, Takifugu rubripes larvae were exposed to four different LED photoperiods: 8 L:16 D, 16 L:8 D, 20 L:4 D and 24 L:0 D from −1 to 30 days post hatching (DPH). We measured growth, activity of pepsin (PPS), lipase (LPS), alanine aminotransferase (ALT), aspartate aminotransferase (AST), and non-specific immune enzyme activity of alkaline phosphatase (AKP), and acid phosphate (ACP), superoxide dismutase (SOD), Catalase (CAT), glutathione peroxide (GSH-PX), total antioxidant capacity (T-AOC) at 1, 5, 10, 15, 20, and 30 DPH. The results showed that there was no difference in the effects of photoperiod on the growth within 18 DHP, while 20 L:4 D had a positive effect on the growth of the larvae after 19 DPH. The effect of photoperiod on digestive enzymes did not show obvious regularity. Compared with photoperiod, the changes of some digestive and metabolic enzymes were mainly affected by exogenous food (LPS, ALT, AST changed obviously after feeding Artemia). On the 30 DPH, the AKP of 8 L: 16 D and 24 L: 0 D groups was significantly higher than that of the other two groups, which indicated that 8 L: 16 D and 24 L: 0 D lighting might cause stress to fish body and lead to the increase of AKP. GSH-PX in newly hatched fish at 24 L:0 D was significantly higher than that in other groups. Photoperiod had no significant effect on the enzyme activity of PPS, LDH, ACP, CAT and no obvious pattern was found. Overall, our results suggested that 20 L:4 D promoted larval growth after 19 DPH and shorter photoperiods can be used to save electricity before 18 DPH of Takifugu rubripes larva.

Cu2+ exhibited a good activation effect on specularite. However, its microscopic activation mechanism needs further study. Additionally, Cu2+ was mainly present in the flotation solution as Cu2+, Cu(OH)+, and Cu(OH)2 at pH = 7. Therefore, density functional theory (DFT) calculations were used to investigate the effect of Cu species such as Cu2+, Cu(OH)+, and Cu(OH)2 adsorbed on the crystal structure and properties of the specularite (0 0 1) surface. The adsorption mechanism of different Cu components on the surface was also further clarified by the analyses of the adsorption model, adsorption energy, partial density of states (PDOS), charge transfer, and bond properties. In addition, the obtained results are discussed. Based on the obtained results, it can be concluded that the geometric structure and electronic properties on the surface changed after adsorbing Cu components and that the O3–Fe1–O1 structure was more susceptible to the adsorbates. The adsorption engines results show that Cu components could be spontaneously adsorbed onto the specularite (0 0 1) surface with adsorption energies of −0.76, −0.85, and −1.78 eV, corresponding to Cu2+, CuOH+, and Cu(OH)2, respectively. Therefore, the adsorption stability of the Cu species on the specularite surface increased in the order of Cu2+ &lt, Cu(OH)+ &lt, Cu(OH)2. Additionally, the adsorption sites for Cu species on the surface were different. Cu2+ interacted mainly with O atoms on the surface, forming Cu–O complexes, while Cu(OH)+ and Cu(OH)2 acted mainly through the O atom of –OH, interacting with Fe atoms to form Cu–O–Fe complexes. The formation of Cu–O and Cu–O–Fe complexes increased the adsorption sites for sodium oleate, with more hydrophobic species being generated to improve the floatability of specularite.

In this paper, a laser triangulation sensor (LTS) with improved measurement accuracy via integrating a diffraction grating was proved. Compared to conventional LTS that only has the information of one light spot, the diffraction grating generates several diffraction light spots on CMOS image sensor. This enables to obtain multiple results of the object displacement simultaneously during one sampling period. Thus, it is achievable to obtain a higher measurement accuracy by averaging operation, due to the averaging-error effect. The operating principle of the proposed method is explained in detail, and a prototype LTS is developed. Contrastive experiments are performed to evaluate the nonlinearity and repeatability of the developed LTS, which are generally combined to present the measurement accuracy of LTS. In the full-scale measurement range of 20 mm (F.S. = 20 mm), the results show that the measurement nonlinearity and repeatability of the developed LTS with diffraction grating are 0.113% of F.S. and 0.89 μm respectively, which are significantly improved compared to that without diffraction grating.

As one of the fundamental problems for image enhancing editing, image cropping seeks the best crop with high aesthetic quality and removes extraneous areas. Many recent deep learning methods have been proposed to address the problem, but they do not reveal the intrinsic mechanism of image cropping. In this paper, we explore the latent region pair and then fulfill its potential in our proposed deep learning methodology (SAIC-Net) with saliency map for automatic image cropping. For each image, a lightweight multi-scale feature extraction network is first adopted to produce deep and informative features. Then, the features of latent region pair (ROI and ROD) are aligned and refined by the proposed saliency-aware align operators and context channel attentions. Finally, hybrid loss composed by ranking loss and Huber loss is minimized when training our model. In our experiments, to reduce the searching space for candidate crops, we conduct a saliency-aware grid cropping candidates generation method to eliminate irrelevant crops. Afterwards, we provide a thorough ablation study to better figure out the superiority of each part in our method, and conduct user study against the state-of-the-art methods on a fraction of performance metrics. The quantitative and qualitative results on three benchmark datasets demonstrate the superiority of our SAIC-Net in the task of automatic cropping.