Does settlement land expansion necessarily induce the decrease of cultivated land? Differences in national scale and local counties of China.

With the extensive industrialization and urbanization taking place in China during the recent decades, land use throughout the country has experienced profound changes influenced not only by the demand for population growth and living standard improvement but also by the constraints of series of land use policies. However, whether the conflict between the expansion of settlement land (SL) and the loss of cultivated land (CL) have been resolved at the national scale or transferred between the local regions remains unclear. Based on yearly ESA CCI land use and land cover products from 1992 to 2020, the long-term trends of quantity and spatial pattern of SL expansion and CL change in China from national and local views were investigated using trend statistic methods, and finally a comprehensive zoning framework was proposed to recognize the trade-off and synergies relationships between SL expansion and CL change. There are a continuous expansion of SL with global linear trends showing three breakpoints in 2000, 2005, and 2012, and a fluctuation decline of CL presented with four breakpoints in 1997, 2002, 2006, and 2013. Aggregation and dispersion tendencies with linear characteristics of SL expansion and CL change were found with breakpoints in 2001, 2008, 2012, and 2016 and breakpoints in 2001 and 2010, respectively. A spotty spatial pattern of SL was shown spatially coincident with urban agglomerations in China while the planar continuous characteristic was found for CL. Local counties were classified into five tradeoff and synergies zones (TSZs), where general synergies (G-S) and decoupling (D) of SL expansion and CL change were rare cases and the different change in quantity and trend of SL expansion and CL change in local counties was concealed by the national trend. A few scattered counties were belonging to G-S and D TSZs, while most of the counties in the central-east and western China were in General-Tradeoff (G-T) and Superior-Tradeoff (S-T) TSZs. Counties in south and north China with higher percentages of CL were more prevalent in Superior-Synergy (S-S) TSZ. Our findings explicated the complex relationships between SL expansion and CL change of China at the national scale and in local counties, which pointed out the differences of unified land use management activities across scales and could provide insights for future policy-making and management measures of land use to both ensure the national food security and promote regional sustainable development more synchronously.

A causal association between amyotrophic lateral sclerosis and atrial fibrillation: a two-sample Mendelian randomization study.

Objectives: To look into the connection between amyotrophic lateral sclerosis (ALS) and atrial fibrillation (AF) using Mendelian randomization (MR). Methods: Two-sample MR was performed using genetic information from genome-wide association studies (GWAS). Genetic variants robustly associated with ALS and AF were used as instrumental variables. GWAS genetic data for ALS (n = 138,086, ncase = 27,205) and AF (n = 1,030,836, ncase = 60,620), publicly available from IEU Open. The specific MR protocols were Inverse variance-weighted (IVW), Simple mode, MR Egger, Weighted mode, and Weight median estimator (WME). Subsequently, the MR-Egger intercept and Cochran Q examine were used to evaluate instrumental variables (IVs)' heterogeneity and multiplicative effects (IVs). In addition, MR-PRESSO analysis was conducted to exclude any potential pleiotropy. Results: The IVW method demonstrated that ALS positively affected AF [OR: 1.062, 95% CI (1.004-1.122); P = 0.035]. Indeed, other MR methods were in accordance with the tendency of the IVW method (all OR > 1), and sensitivity testing verified the reliability of this MR result. Conclusions: This MR study proves a positive causal connection between ALS and atrial fibrillation. Further studies are warranted to elucidate the mechanisms linking ALS and AF.

Dual effects on vegetation from urban expansion in the drylands of northern China: A multiscale investigation using the vegetation disturbance index.

Drylands contribute roughly 40 % of the global net primary productivity and are essential for achieving sustainable development. Investigating the effects on vegetation from urban expansion in drylands within the context of rapid urbanization could help enhance the sustainability of dryland cities. With the use of the drylands of northern China (DNC) as an example, we applied the vegetation disturbance index to investigate the negative and positive effects on vegetation from urban expansion in drylands. The results revealed that the DNC experienced massive and rapid urban expansion from 2000 to 2020. Urban land in the entire DNC increased by 19,646 km2 from 8141 to 27,787 km2, with an annual growth rate of 6.3 %. Urban expansion in the DNC imposed both negative and positive effects on regional vegetation. The area with negative effects reached 7736 km2 and was mainly concentrated in the dry subhumid zones. The area with positive effects amounted to 5011 km2 and was comparable among the dry subhumid, semiarid, and arid zones. Land use/cover change induced by population growth significantly contributed to these negative effects, while the positive effects were largely caused by economic growth. Therefore, it is recommended to strike a balance between urban growth and vegetation conservation to mitigate the adverse effects on vegetation from urban expansion in drylands. Simultaneously, it is imperative to expand urban green spaces and build sustainable and livable ecological cities to facilitate sustainable urban development.

Astragaloside IV attenuates renal tubule injury in DKD rats via suppression of CD36-mediated NLRP3 inflammasome activation.

Background: In recent years, diabetic kidney disease (DKD) has emerged as a prominent factor contributing to end-stage renal disease. Tubulointerstitial inflammation and lipid accumulation have been identified as key factors in the development of DKD. Earlier research indicated that Astragaloside IV (AS-IV) reduces inflammation and oxidative stress, controls lipid accumulation, and provides protection to the kidneys. Nevertheless, the mechanisms responsible for its protective effects against DKD have not yet been completely elucidated. Purpose: The primary objective of this research was to examine the protective properties of AS-IV against DKD and investigate the underlying mechanism, which involves CD36, reactive oxygen species (ROS), NLR family pyrin domain containing 3 (NLRP3), and interleukin-1ß (IL-1ß). Methods: The DKD rat model was created by administering streptozotocin along with a high-fat diet. Subsequently, the DKD rats and palmitic acid (PA)-induced HK-2 cells were treated with AS-IV. Atorvastatin was used as the positive control. To assess the therapeutic effects of AS-IV on DKD, various tests including blood sugar levels, the lipid profile, renal function, and histopathological examinations were conducted. The levels of CD36, ROS, NLRP3, Caspase-1, and IL-1ß were detected using western blot analysis, PCR, and flow cytometry. Furthermore, adenovirus-mediated CD36 overexpression was applied to explore the underlying mechanisms through in vitro experiments. Results: In vivo experiments demonstrated that AS-IV significantly reduced hyperglycemia, dyslipidemia, urinary albumin excretion, and serum creatinine levels in DKD rats. Additionally, it improved renal structural abnormalities and suppressed the expression of CD36, NLRP3, IL-1ß, TNF-α, and MCP-1. In vitro experiments showed that AS-IV decreased CD36 expression, lipid accumulation, and lipid ROS production while inhibiting NLRP3 activation and IL-1ß secretion in PA-induced HK-2 cells. Conclusion: AS-IV alleviated renal tubule interstitial inflammation and tubule epithelial cell apoptosis in DKD rats by inhibiting CD36-mediated lipid accumulation and NLRP3 inflammasome activation.

Lightweight model-based sheep face recognition via face image recording channel.

The accurate identification of individual sheep is a crucial prerequisite for establishing digital sheep farms and precision livestock farming. Currently, deep learning technology provides an efficient and non-contact method for sheep identity recognition. In particular, convolutional neural networks can be used to learn features of sheep faces to determine their corresponding identities. However, the existing sheep face recognition models face problems such as large model size, and high computational costs, making it difficult to meet the requirements of practical applications. In response to these issues, we introduce a lightweight sheep face recognition model called YOLOv7-Sheep Face Recognition (YOLOv7-SFR). Considering the labor-intensive nature associated with manually capturing sheep face images, we developed a face image recording channel to streamline the process and improve efficiency. This study collected facial images of 50 Small-tailed Han sheep through a recording channel. The experimental sheep ranged in age from 1 to 3 yr, with an average weight of 63.1 kg. Employing data augmentation methods further enhanced the original images, resulting in a total of 22,000 sheep face images. Ultimately, a sheep face dataset was established. To achieve lightweight improvement and improve the performance of the recognition model, a variety of improvement strategies were adopted. Specifically, we introduced the shuffle attention module into the backbone and fused the Dyhead module with the model's detection head. By combining multiple attention mechanisms, we improved the model's ability to learn target features. Additionally, the traditional convolutions in the backbone and neck were replaced with depthwise separable convolutions. Finally, leveraging knowledge distillation, we enhanced its performance further by employing You Only Look Once version 7 (YOLOv7) as the teacher model and YOLOv7-SFR as the student model. The training results indicate that our proposed approach achieved the best performance on the sheep face dataset, with a mean average precision@0.5 of 96.9%. The model size and average recognition time were 11.3 MB and 3.6 ms, respectively. Compared to YOLOv7-tiny, YOLOv7-SFR showed a 2.1% improvement in mean average precision@0.5, along with a 5.8% reduction in model size and a 42.9% reduction in average recognition time. The research results are expected to drive the practical applications of sheep face recognition technology.

Accurate identification of individual sheep is a crucial prerequisite for establishing digital sheep farms and precision livestock farming. In this study, we developed a lightweight sheep face recognition model, YOLOv7-SFR. Utilizing a face image recording channel, we efficiently collected facial images from 50 experimental sheep, resulting in a comprehensive sheep face dataset. Training results demonstrated that YOLOv7-SFR surpassed state-of-the-art lightweight sheep face recognition models, achieving a mean average precision@0.5 of 96.9%. Notably, the model size and average recognition time of YOLOv7-SFR were merely 11.3 MB and 3.6 ms, respectively. In summary, YOLOv7-SFR strikes an optimal balance between performance, model size, and recognition speed, offering promising practical applications for sheep face recognition technology. This study employs deep learning for sheep face recognition tasks, ensuring the welfare of sheep in the realm of digital agriculture and automation practices.

scSemiGCN: boosting cell-type annotation from noise-resistant graph neural networks with extremely limited supervision.

MOTIVATION: Cell-type annotation is fundamental in revealing cell heterogeneity for single-cell data analysis. Although a host of works have been developed, the low signal-to-noise-ratio single-cell RNA-sequencing data that suffers from batch effects and dropout still poses obstacles in discovering grouped patterns for cell types by unsupervised learning and its alternative-semi-supervised learning that utilizes a few labeled cells as guidance for cell-type annotation. RESULTS: We propose a robust cell-type annotation method scSemiGCN based on graph convolutional networks. Built upon a denoised network structure that characterizes reliable cell-to-cell connections, scSemiGCN generates pseudo labels for unannotated cells. Then supervised contrastive learning follows to refine the noisy single-cell data. Finally, message passing with the refined features over the denoised network structure is conducted for semi-supervised cell-type annotation. Comparison over several datasets with six methods under extremely limited supervision validates the effectiveness and efficiency of scSemiGCN for cell-type annotation. AVAILABILITY AND IMPLEMENTATION: Implementation of scSemiGCN is available at https://github.com/Jane9898/scSemiGCN.

Ice Adhesion Properties on Micropillared Superhydrophobic Surfaces.

In this work, we investigate the freezing behavior and ice adhesion properties of sessile drops on micropillared superhydrophobic surfaces (SHSs) with various sizes, which are of practical importance for anti/deicing. First of all, it is demonstrated that the recalescence is related only to the supercooling degree of drops but not to the geometrical parameters of micropillars. The freezing time of sessile drops first increases and then decreases with the area fraction of the SHSs, which demonstrates the nonmonotonic dependence of the icing time on the area fraction. Moreover, the influence of the geometrical parameters of the micropillars on the ice adhesion is discussed. With the decrease of the substrate temperature, the wetting state of the adhesive ice can be transformed from the Cassie ice to the Wenzel ice. For the Cassie ice, the adhesive force is proportional to the area fraction of the SHSs. Interestingly, experimental results show that there exist two interfacial debonding modes of the Wenzel ice: translational debonding and rotational debonding. Furthermore, it is found that the rotational debonding mode contributes to a much lower adhesive force between the ice and the micropillared surface compared to that of the translational debonding mode. By analyzing the critical interfacial energy release rate of the two modes, we deduce the threshold between the two modes, which is quantified as the geometrical parameters of the micropillars. In addition, quantitative relations between the geometrical parameters and the adhesion strengths of the two modes are also obtained. We envision that this work would shed new light on the design optimization of anti/deicing materials.

Ofatumumab subcutaneous injection successfully treated patients with pemphigus vulgaris relapse post rituximab.

The therapeutic strategy for the treatment of pemphigus vulgaris (PV) still needs optimization because of the multiple deficiencies of glucocorticoid and rituximab. Ofatumumab, another CD20 monoclonal antibody administrated subcutaneously, provides a possible alternative option. In this study, three patients experienced PV relapse after clinical remission induced by rituximab. With written informed consent, they received an ofatumumab (20 mg) subcutaneous injection twice (2 weeks apart) in combination with a prednisone dose adjusted according to their weight and disease severity. Over the 24-week observation, two of three patients achieved lesion clear-up under prednisone (0.2 mg/kg per day), and the other patient's pemphigus disease area index dropped from 39 to 3 with prednisone (15 mg/day). The anti-desmoglein antibody levels and CD19+ B cell counts declined compared to those at baseline. No severe adverse events were observed within the 24-week follow-up. In summary, we propose a protocol of ofatumumab for patients with refractory PV and report positive treatment outcomes of three patients who received this regimen.

Emodin alleviates CRS4-induced mitochondrial damage via activation of the PGC1α signaling.

Cardiorenal syndrome type 4 (CRS4), a progressive deterioration of cardiac function secondary to chronic kidney disease (CKD), is a leading cause of death in patients with CKD. In this study, we aimed to investigate the cardioprotective effect of emodin on CRS4. C57BL/6 mice with 5/6 nephrectomy and HL-1 cells stimulated with 5% CKD mouse serum were used for in vivo and in vitro experiments. To assess the cardioprotective potential of emodin, we employed a comprehensive array of methodologies, including echocardiography, tissue staining, immunofluorescence staining, biochemical detection, flow cytometry, real-time quantitative PCR, and western blot analysis. Our results showed that emodin exerted protective effects on the function and structure of the residual kidney. Emodin also reduced pathologic changes in the cardiac morphology and function of these mice. These effects may have been related to emodin-mediated suppression of reactive oxygen species production, reduction of mitochondrial oxidative damage, and increase of oxidative metabolism via restoration of PGC1α expression and that of its target genes. In contrast, inhibition of PGC1α expression significantly reversed emodin-mediated cardioprotection in vivo. In conclusion, emodin protects the heart from 5/6 nephrectomy-induced mitochondrial damage via activation of the PGC1α signaling. The findings obtained in our study can be used to develop effective therapeutic strategies for patients with CRS4.

Undersized telomeres in regulatory T cells link to the pathogenesis of allergic rhinitis.

Telomeres are an important biomarker in the cell destiny. The relationship between telomeres and regulatory T cells (Tregs) has not yet been investigated. The objective of this study is to evaluate the link between Tregs' telomere length and allergic rhinitis (AR)'s pathogenesis. Here, we report that low telomerase activity and high endoplasmic reticulum stress status were observed in Tregs from AR patients, as shown in the results. Immune regulatory molecules levels were correlated with the length of Tregs' telomeres. The immune-suppressive functions of Tregs were associated with the telomere length/Telomerase reverse transcriptase/Telomerase protein component 1 status in Tregs. The levels of telomere length/telomerase in airway Tregs were reduced by sensitization. Endoplasmic reticulum stress signaling pathway of proline-rich receptor-like protein kinase-eukaryotic translation initiation factor 2A (eIF2a) was associated with the regulation of telomerase. Inhibiting eIF2a had an effect on upregulating telomerase activity in Tregs and mitigating experimental AR.

Mesoscopic Model for Reversible Adsorption Stage of Albumin and Fibrinogen on TiO2 Surface.

The competitive behavior of proteins in the reversible adsorption stage plays a crucial role in determining the composition of the protein layer and the subsequent biological responses to the biomaterial. However, such competitive adsorption is a mesoscopic process at physiological protein concentration, and neither a macroscopic experiment nor microscopic MD (molecular dynamics) simulation is suitable to clarify it. Here, we proposed a mesoscopic DPD (dissipative particle dynamics) model to illustrate the competitive process of albumin and fibrinogen on TiO2 surface with its parameters deduced from our previous MD simulation, and proved the model well retained the diffusion and adsorption properties of proteins in the competitive adsorption on the plane surface. We then applied the model to the competitive adsorption on the surfaces with different nanostructures and observed that when the nanostructure size is much larger than that of protein, the increase in surface area is the main influencing factor; when the nanostructure size is close to that of protein, the coordination between the nanostructure and the size and shape of protein significantly affects the competitive adsorption process. The model has revealed many mechanical phenomena observed in previous experimental studies and has the potential to contribute to the development of high-performance biomaterials.

Fully Electrically Controlled van der Waals Multiferroic Tunnel Junctions.

The fully electrical control of the magnetic states in magnetic tunnel junctions is highly pursued for the development of the next generation of low-power and high-density information technology. However, achieving this functionality remains a formidable challenge at present. Here we propose an effective strategy by constructing a trilayer van der Waals multiferroic structure, consisting of CrI3-AgBiPSe6 and Cr2Ge2Te6-In2Se3, to achieve full-electrical control of multiferroic tunnel junctions. Within this structure, two different magnetic states of the magnetic bilayers (CrI3/Cr2Ge2Te6) can be modulated and switched in response to the polarization direction of the adjacent ferroelectric materials (AgBiPSe6/In2Se3). The intriguing magnetization reversal is mainly attributed to the polarization-field-induced band structure shift and interfacial charge transfer. On this basis, we further design two multiferroic tunnel junction devices, namely, graphene/CrI3-AgBiPSe6/graphene and graphene/Cr2Ge2Te6-In2Se3/graphene. In these devices, good interfacial Ohmic contacts are successfully obtained between the graphene electrode and the heterojunction, leading to an ultimate tunneling magnetoresistance of 9.3 × 106%. This study not only proposes a feasible strategy and identifies a promising candidate for achieving fully electrically controlled multiferroic tunnel junctions but also provides insights for designing other advanced spintronic devices.

Transcriptome-based network analysis of cell cycle-related genes in response to blue and red light in maize.

In maize, blue and red light are key environmental factors regulating cell-cycle progression. We used transcriptomics to investigate and compare differential gene expression under the four light conditions: red light, blue light, red converted to blue and blue converted to red. A total of 23 differentially expressed genes were identified. The gene-gene interaction analysis indicated a significant interaction between four unidentified genes, 100191551, pco143873, 100284747 and pco060490, and cell-cycle-related genes. Using multiple sequence alignment analysis and protein structure comparisons, we show here that these four unidentified genes were characterized as ALP1-like, ALP1, cyclin P1-1 and AEBP2, respectively. By constructing a protein-protein interaction network, we inferred that 100191551 and pco143873 are potentially regulated to avoid DNA damage by abiotic stress response factors in the cell cycle. The gene 100284747 regulates the cell cycle in response to phosphate starvation signalling. The gene pco060490 potentially negatively regulates the cell cycle through the mediation of Histone H3 and CYCD6 in response to red light. In conclusion, the cell-cycle-related genes are sensitive to blue and red light, and four novel functional genes may be involved in the cell cycle.

Diurnal oscillations of epigenetic modifications are associated with variation in rhythmic expression of homoeologous genes in Brassica napus.

BACKGROUND: Epigenetic modifications that exhibit circadian oscillations also promote circadian oscillations of gene expression. Brassica napus is a heterozygous polyploid species that has undergone distant hybridization and genome doubling events and has a young and distinct species origin. Studies incorporating circadian rhythm analysis of epigenetic modifications can offer new insights into differences in diurnal oscillation behavior among subgenomes and the regulation of diverse expressions of homologous gene rhythms in biological clocks. RESULTS: In this study, we created a high-resolution and multioscillatory gene expression dataset, active histone modification (H3K4me3, H3K9ac), and RNAPII recruitment in Brassica napus. We also conducted the pioneering characterization of the diurnal rhythm of transcription and epigenetic modifications in an allopolyploid species. We compared the evolution of diurnal rhythms between subgenomes and observed that the Cn subgenome had higher diurnal oscillation activity in both transcription and active histone modifications than the An subgenome. Compared to the A subgenome in Brassica rapa, the An subgenome of Brassica napus displayed significant changes in diurnal oscillation characteristics of transcription. Homologous gene pairs exhibited a higher proportion of diurnal oscillation in transcription than subgenome-specific genes, attributed to higher chromatin accessibility and abundance of active epigenetic modification types. We found that the diurnal expression of homologous genes displayed diversity, and the redundancy of the circadian system resulted in extensive changes in the diurnal rhythm characteristics of clock genes after distant hybridization and genome duplication events. Epigenetic modifications influenced the differences in the diurnal rhythm of homologous gene expression, and the diurnal oscillation of homologous gene expression was affected by the combination of multiple histone modifications. CONCLUSIONS: Herein, we presented, for the first time, a characterization of the diurnal rhythm characteristics of gene expression and its epigenetic modifications in an allopolyploid species. Our discoveries shed light on the epigenetic factors responsible for the diurnal oscillation activity imbalance between subgenomes and homologous genes' rhythmic expression differences. The comprehensive time-series dataset we generated for gene expression and epigenetic modifications provides a valuable resource for future investigations into the regulatory mechanisms of protein-coding genes in Brassica napus.

The numerical simulation of a four-stage linear transformer driver module based on the Preisach magnetic core model.

The use of linear transformer drivers (LTDs) is widely considered the most promising technological approach for the development of future pulsed-power accelerators. In large-scale pulsed-power accelerators, abnormal conditions like switch prefire can occur frequently during tests and normal operations due to the presence of a large number of switches. The diagnosis of such faults based on signature waveforms requires further investigation. According to previous research, the characteristics of the magnetic cores greatly influence the fault waveforms. In this paper, a full-cycle mathematical model of the magnetic core is established utilizing a classical Preisach model based on experimental results. This model is coupled with the LTD circuit model in simulations, and simulation results are obtained under the condition of switch prefire. The simulation results are in good agreement with the experimental results from a four-stage LTD module with a sharing shell and de-ionized water insulated transmission line. The magnetization process of the cores is also determined under prefire conditions. Analyses of the magnetization process indicate that the completely demagnetized core shows high permeability under positive excitation and that the permeability abruptly decreases as the excitation is reversed. The hysteresis characteristics result in a phenomenon in which the output voltage in the prefired stage is almost unipolar. Finally, the features of the fault voltages captured in the experiments are also explained.

SerpinA3 Promotes Myocardial Infarction in Rat and Cell-based Models.

This study aimed to examine the role and molecular mechanism of the nuclear factor κB (NFκB)/serine protease inhibitor A3 (SerpinA3) interaction in myocardial ischemia-reperfusion (IR) injury. First, a rat model for myocardial ischemia-reperfusion injury was established, using 2,3,5-triphenyltetrazolium chloride to measure the size of the myocardial infarction. Pathological variations in myocardial tissue were detected using hematoxylin-eosin staining. Flow cytometry and terminal deoxynucleotidyl transferase biotin-dUTP nick end labeling (TUNEL) staining were used to measure cell death in the rat model. The SerpinA3 mRNA and protein expressions in the myocardium of IR-model rats were remarkably higher than those in the control group. Furthermore, the oxidative, inflammatory, and apoptotic activities of the myocardial tissue of SerpinA3-knockdown (KD) rats were significantly improved compared to those in the WT group. SerpinA3-KD also contributed to the recovery of cardiac function in IR-model rats. Additionally, silencing of SerpinA3 inhibited p65 phosphorylation in myocardial tissues and reduced H2O2-induced inflammation, oxidative stress, and apoptosis in myocardial cells. The expression of SerpinA3 increased in myocardial tissue after IR stimulation. Knockdown of SerpinA3 can deactivate NF-κB and reduce inflammation, oxidative stress, and apoptosis in vivo and in vitro, thereby lessening myocardial injury caused by IR. In conclusion, SerpinA3 promotes myocardial infarction in rat and cell-based models by activating NF-κB. However, the mechanism by which increased Serpina3 expression causes downstream NF-κB activation to mediate the proposed, pathological effects in myocardial IR injury remain untested and worthy of future investigations.

LSR-YOLO: A High-Precision, Lightweight Model for Sheep Face Recognition on the Mobile End.

The accurate identification of sheep is crucial for breeding, behavioral research, food quality tracking, and disease prevention on modern farms. As a result of the time-consuming, expensive, and unreliable problems of traditional sheep-identification methods, relevant studies have built sheep face recognition models to recognize sheep through facial images. However, the existing sheep face recognition models face problems such as high computational costs, large model sizes, and weak practicality. In response to the above issues, this study proposes a lightweight sheep face recognition model named LSR-YOLO. Specifically, the ShuffleNetv2 module and Ghost module were used to replace the feature extraction module in the backbone and neck of YOLOv5s to reduce floating-point operations per second (FLOPs) and parameters. In addition, the coordinated attention (CA) module was introduced into the backbone to suppress non-critical information and improve the feature extraction ability of the recognition model. We collected facial images of 63 small-tailed Han sheep to construct a sheep face dataset and further evaluate the proposed method. Compared to YOLOv5s, the FLOPs and parameters of LSR-YOLO decreased by 25.5% and 33.4%, respectively. LSR-YOLO achieved the best performance on the sheep face dataset, and the mAP@0.5 reached 97.8% when the model size was only 9.5 MB. The experimental results show that LSR-YOLO has significant advantages in recognition accuracy and model size. Finally, we integrated LSR-YOLO into mobile devices and further developed a recognition system to achieve real-time recognition. The results show that LSR-YOLO is an effective method for identifying sheep. The method has high recognition accuracy and fast recognition speed, which gives it a high application value in mobile recognition and welfare breeding.