Search Results (3,061)

Zinc (Zn) deficiency poses a major global health challenge, and wheat grains generally contain low Zn concentrations. In this study, the wheat cultivar ‘Zhongmai 175’ was identified as zinc-efficient. Hydroponic experiments demonstrated that Zn deficiency induced the secretion of oxalic acid and malic acid in root exudates and significantly increased total root length in ‘Zhongmai 175’. To elucidate the underlying regulatory mechanisms, transcriptome profiling via RNA sequencing was conducted under Zn-deficient conditions. A total of 2287 and 1935 differentially expressed genes (DEGs) were identified in roots and shoots, respectively. Gene Ontology enrichment analysis revealed that these DEGs were primarily associated with Zn ion transport, homeostasis, transmembrane transport, and hormone signaling. Key DEGs belonged to gene families including VIT, NAS, DMAS, ZIP, tDT, HMA, and NAAT. KEGG pathway analysis indicated that phenylpropanoid biosynthesis, particularly lignin synthesis genes, was significantly downregulated in Zn-deficient roots. In shoots, cysteine and methionine metabolism, along with plant hormone signal transduction, were the most enriched pathways. Notably, most DEGs in shoots were associated with the biosynthesis of phytosiderophores (MAs, NA) and ethylene. Overall, genes involved in Zn ion transport, phytosiderophore biosynthesis, dicarboxylate transport, and ethylene biosynthesis appear to play central roles in wheat’s adaptive response to Zn deficiency. These findings provide a valuable foundation for understanding the molecular basis of Zn efficiency in wheat and for breeding Zn-enriched varieties. Full article

Food security is crucial for social stability and economic development. Ensuring food security in the arid region of Northwest China presents unique challenges due to limited water and soil resources. This study addresses these challenges by integrating a comprehensive water and soil resource matching assessment with grain production forecasting. Based on data from 2000 to 2020, this research projects the food security status to 2035 using the GM(1,1) model, incorporating a comprehensive index of soil and water resource matching and regression analysis to inform production forecasts. Key assumptions include continued historical trends in population growth, urbanization, and dietary shifts towards an increased animal protein consumption. The findings revealed a consistent upward trend in grain production from 2000 to 2020, with an average annual growth rate of 3.5%. Corn and wheat emerged as the dominant grain crops. Certain provinces demonstrated comparative advantages for specific crops like rice and wheat. The most significant finding is that despite the projected growth in the total grain output by 2035 compared to 2020, the regional grain self-sufficiency rate is projected to range from 79.6% to 84.1%, falling below critical food security benchmarks set by the FAO and China. This projected shortfall carries significant implications, underscoring a serious challenge to regional food security and highlighting the region’s increasing vulnerability to external food supply fluctuations. The findings strongly signal that current trends are insufficient and necessitate urgent and proactive policy interventions. To address this, practical policy recommendations include promoting water-saving technologies, enhancing regional cooperation, and strategically utilizing the international grain trade to ensure regional food security. Full article

Water scarcity and spatial variability in soil fertility are key constraints to stable grain production in the Huang-Huai-Hai Plain. However, the interaction mechanisms between regulated deficit irrigation and soil fertility influencing yield formation and water-nitrogen use efficiency in winter wheat remain unclear. In this study, a two-year field experiment (2022–2024) was conducted to investigate the effects of two irrigation regimes—regulated deficit irrigation during the heading to grain filling stage (D) and full irrigation (W)—under four soil fertility levels: F1 (N: P: K = 201.84: 97.65: 199.05 kg ha⁻¹), F2 (278.52: 135: 275.4 kg ha⁻¹), F3 (348.15: 168.75: 344.25 kg ha⁻¹), and CK (no fertilization). The results show that aboveground dry matter accumulation, total nitrogen content, pre-anthesis dry matter and nitrogen translocation, and post-anthesis accumulation significantly increased with fertility level (p < 0.05). Regulated deficit irrigation promoted the contribution of post-anthesis dry matter to grain yield under the CK and F1 treatments, but suppressed it under the F2 and F3 treatments. However, it consistently enhanced the contribution of post-anthesis nitrogen to grain yield (p < 0.05) across all fertility levels. Higher fertility levels prolonged the grain filling duration by 18.04% but reduced the mean grain filling rate by 15.05%, whereas regulated deficit irrigation shortened the grain filling duration by 3.28% and increased the mean grain filling rate by 12.83% (p < 0.05). Grain yield significantly increased with improved fertility level (p < 0.05), reaching a maximum of 9361.98 kg·ha⁻¹ under the F3 treatment. Regulated deficit irrigation increased yield under the CK and F1 treatments but reduced it under the F2 and F3 treatments. Additionally, water use efficiency exhibited a parabolic response to fertility level and was significantly enhanced by regulated deficit irrigation. Nitrogen partial factor productivity (NPFP) declined with increasing fertility level (p < 0.05); Regulated deficit irrigation improved NPFP under the F1 treatment but reduced it under the F2 and F3 treatments. The highest NPFP (41.63 kg·kg⁻¹) was achieved under the DF1 treatment, which was 54.81% higher than that under the F3 treatment. TOPSIS analysis showed that regulated deficit irrigation combined with the F1 fertility level provided the optimal balance among yield, WUE, and NPFP. Therefore, implementing regulated deficit irrigation during the heading–grain filling stage under moderate fertility (F1) is recommended as the most effective strategy for achieving high yield and efficient resource utilization in winter wheat production in this region. Full article

In the irrigation districts of Northern China, the flood resources utilization for deep storage irrigation, which is essentially characterized by active excessive irrigation, aims to have the potential to mitigate freshwater shortages, and long-term groundwater overexploitation. It is crucial to detect the effects of irrigation amounts on agricultural yield and the mechanisms under deep storage irrigation. A three-year field experiment (2020–2023) was conducted in the Guanzhong Plain, according to five soil wetting layer depths (RF: 0 cm; W1: control, 120 cm; W2: 140 cm; W3: 160 cm; W4: 180 cm) with soil saturation water content as the irrigation upper limit. Results exhibited that, compared to W1, the W2, W3, and W4 treatments led to the increased plant height, leaf area index, and dry matter accumulation. Meanwhile, the W2, W3, and W4 treatments improved kernel weight increment achieving maximum grain-filling rate (W_max), maximum grain-filling rate (G_max), and average grain-filling rate (G_ave), thereby enhancing the effective spikes (ES) and grain number per spike (GS), and thus increased wheat grain yield (GY). In relative to W1, the W2, W3, and W4 treatments increased the ES, GS, and GY by 11.89–19.81%, 8.61–14.36%, and 8.17–13.62% across the three years. Notably, no significant difference was observed in GS and GY between W3 and W4 treatments, but W4 treatment displayed significant decreases in ES by 3.04%, 3.06%, and 2.98% in the respective years. The application of a structural equation modeling (SEM) revealed that deep storage irrigation improved ES and GS by positively regulating W_max, G_max, and G_ave, thus significantly increasing GY. Overall, this study identified the optimal threshold (W3 treatment) to maximize wheat yields by optimizing both the vegetative growth and grain-filling dynamics. This study provides essential support for the feasibility assessment of deep storage irrigation before flood seasons, which is vital for the balance and coordination of food security and water security. Full article

Wheat is one of the important grain crops, and spike counting is crucial for predicting spike yield. However, in complex farmland environments, the wheat body scale has huge differences, its color is highly similar to the background, and wheat ears often overlap with each other, which makes wheat ear detection work face a lot of challenges. At the same time, the increasing demand for high accuracy and fast response in wheat spike detection has led to the need for models to be lightweight function with reduced the hardware costs. Therefore, this study proposes a lightweight wheat ear detection model, CML-RTDETR, for efficient and accurate detection of wheat ears in real complex farmland environments. In the model construction, the lightweight network CSPDarknet is firstly introduced as the backbone network of CML-RTDETR to enhance the feature extraction efficiency. In addition, the FM module is cleverly introduced to modify the bottleneck layer in the C2f component, and hybrid feature extraction is realized by spatial and frequency domain splicing to enhance the feature extraction capability of wheat to be tested in complex scenes. Secondly, to improve the model’s detection capability for targets of different scales, a multi-scale feature enhancement pyramid (MFEP) is designed, consisting of GHSDConv, for efficiently obtaining low-level detail information and CSPDWOK for constructing a multi-scale semantic fusion structure. Finally, channel pruning based on Layer-Adaptive Magnitude Pruning (LAMP) scoring is performed to reduce model parameters and runtime memory. The experimental results on the GWHD2021 dataset show that the ${\mathrm{AP}}_{50}$ of CML-RTDETR reaches 90.5%, which is an improvement of 1.2% compared to the baseline RTDETR-R18 model. Meanwhile, the parameters and GFLOPs have been decreased to 11.03 M and 37.8 G, respectively, resulting in a reduction of 42% and 34%, respectively. Finally, the real-time frame rate reaches 73 fps, significantly achieving parameter simplification and speed improvement. Full article

The production of winter wheat, spring barley, spring oilseed rape, and field beans requires detailed experimental data studies to analyze the quality and productivity of spring barley grain under different cultivation and tillage conditions. As the world’s population grows, more food is required to maintain a stable food supply chain. For many years, intensive farming systems have been used to meet this need. Today, intensive climate change events and other global environmental challenges are driving a shift towards sustainable use of natural resources and simplified cultivation methods that produce high-quality and productive food. It is important to study different tillage systems in order to understand how these methods can affect the chemical composition and nutritional value of the grain. Both agronomic and economic aspects contribute to the complexity of this field and their analysis will undoubtedly contribute to the development of more efficient agricultural practice models and the promotion of more conscious consumption. An appropriate tillage system should be oriented towards local climatic characteristics and people’s needs. The impact of reduced tillage on these indicators in spring barley production is still insufficiently investigated and requires further analysis at a global level. This study was carried out at Vytautas Magnus University Agriculture Academy (Lithuania) in 2022–2024. Treatments were arranged using a split-plot design. Based on a long-term tillage experiment, five tillage systems were tested: deep and shallow plowing, deep cultivation–chiseling, shallow cultivation–disking, and no-tillage. The results show that in 2022–2024, the hectoliter weight and moisture content of spring barley grains increased, but protein content and germination decreased in shallowly plowed fields. In deep cultivation–chiseling fields, the protein content (0.1–1.1%) of spring barley grains decreased, and in shallow cultivation–disking fields, the moisture content (0.2–0.3%) decreased. In all fields, the simplified tillage systems applied reduced spring barley germination (0.4–16.7%). Tillage systems and meteorological conditions are the two main forces shaping the quality indicators of spring barley grains. Properly selected tillage systems and favorable climatic conditions undoubtedly contribute to better grain properties and higher yields, while reducing the risk of disease spread. Full article

Simulation is an important technical tool in corn threshing operations, and the establishment of the corn kernel model is the core part of the simulation process. The existing modeling method is to treat the whole kernel as a rigid body, which cannot be crushed during the simulation process, and the calculation of the crushing rate needs to be considered through multiple criteria such as the contact force, the number of collisions, and so on. Aiming at the issue that kernel crushing during maize threshing cannot be accurately modeled in discrete element simulations, in this study, a sub-area crushing model was constructed; representative samples with 26%, 30% and 34% moisture content were selected from a double-season maturing region in China; based on the physical dimensions and biological structure of the maize kernel, three stress regions were defined; and mechanical property tests were conducted on each of the three stress regions using a texturometer as a way to determine the different crushing forces due to the heterogeneity of the maize structure. The correctness of the model was verified by stacking angle and mechanical property experiments. A discrete element model of corn kernels was established using the Bonding V2 method and sub-area modeling. Bonding parameters were calculated by combining stacking angle tests and mechanical property tests. The flattened corn kernel was used as a prototype, and the bonding parameters were determined through size and mechanical property tests. A 22-ball bonding model was developed using dimensional parameters, and the kernel density was recalculated. Results showed that the relative error between the stacking angle test and the measured mean value was 0.31%. The maximum deviation of axial compression simulation results from the measured mean value was 22.8 N, and the minimum deviation was 3.67 N. The errors between simulated and actual rupture forces at the three force areas were 5%, 10%, and 0.6%, respectively. The decreasing trend of the maximum rupture force for the three moisture levels in the simulation matched that of the actual rupture force. The discrete element model can accurately reflect the rupture force, energy relationship, and rupture process on both sides, top, and bottom of the grain, and it can solve the error problem caused by the contact between the threshing element and the grain line in the actual threshing process to achieve the design optimization of the threshing drum. The modeling method provided in this study can also be applied to breakable discrete element models for wheat and soybean, and it provides a reference for optimizing the design of subsequent threshing devices. Full article

Analyzing the soil organic carbon (SOC) stock in dryland areas of southern Shanxi, particularly under the influence of fertilization and mulching conditions, is crucial for enhancing soil fertility and crop productivity and understanding the SOC pool’s resilience to future climate change scenarios in the region. In a long-term experimental site located in Hongtong County, Shanxi Province, soil samples were collected from the 0–100 cm depth over a nine-year period. These samples were analyzed to evaluate the impact of five treatments: no fertilization and no mulching (CK), conventional farming practices (FP), nitrogen reduction and controlled fertilization (MF), nitrogen reduction and controlled fertilization with ridge-film furrow-sowing (RF), and nitrogen reduction and controlled fertilization with flat-film hole-sowing (FH). The average annual yield of wheat grain, SOC stock, water-soluble organic carbon (WSOC), particulate organic carbon (POC), light fraction organic carbon (LFOC), mineral-associated organic carbon (MOC), and heavy fraction organic carbon (HFOC) stocks were measured. The results revealed that the FH treatment not only significantly increased wheat grain yield but also significantly elevated the SOC stock by 23.71% at the 0–100 cm depth compared to CK. Furthermore, this treatment significantly enhanced the POC, LFOC, and MOC stocks by 106.43–292.98%, 36.93–158.73%, and 17.83–81.55%, respectively, within 0–80 cm. However, it also significantly decreased the WSOC stock by 34.32–42.81% within the same soil layer and the HFOC stock by 72.05–101.51% between the 20 and 100 cm depth. Notably, the SOC stock at the 0–100 cm depth was primarily influenced by the HFOC. Utilizing the DNDC (denitrification–decomposition) model, we found that future temperature increases are detrimental to SOC sequestration in dryland areas, whereas reduced rainfall is beneficial. The simulation results indicated that in a warmer climate, a 2 °C temperature increase would result in a SOC stock decrease of 0.77 to 1.01 t·ha⁻¹ compared to a 1 °C increase scenario. Conversely, under conditions of reduced precipitation, a 20% rainfall reduction would lead to a SOC stock increase of 1.53% to 3.42% compared to a 10% decrease scenario. In conclusion, the nitrogen reduction and controlled fertilization with flat-film hole-sowing (FH) treatment emerged as the most effective practice for increasing SOC sequestration in dryland areas by enhancing the HFOC stock. This treatment also fortified the SOC pool’s capacity to withstand future climate change, thereby serving as the optimal approach for concurrently enhancing production and fertility in this region. Full article

The reduced N fertilization rate and N supplied by pea (Pisum sativum L.) residue may sustain subsequent spring wheat (Triticum aestivum L.) growth, yield, and quality. We examined the response of spring wheat growth, yield, and quality to cropping systems and N fertilization rates from 2012 to 2019 in the US northern Great Plains. Cropping systems were conventional till spring wheat–fallow (CTWF), no-till spring wheat–fallow (NTWF), no-till spring wheat–pea (NTWP), and no-till continuous wheat (NTCW), and N fertilization rates to spring wheat were 0, 50, 100, and 150 kg N ha⁻¹. Wheat plant density and straw yield were 13–100% greater for CTWF and NTWF than NTWP and NTCW in most years. Wheat grain yield and protein concentration were also 15–115% greater for CTWF and NTWF than other cropping systems at most N fertilization rates and years. In contrast, wheat grain test weight was 1–2% lower for CTWF and NTWF at most N fertilization rates and years. Increasing N fertilization rate mostly increased grain yield and protein concentration but reduced grain test weight for most cropping systems and years. Although CTWF and NTWF with or without N fertilization increased wheat yield and quality, these practices are not sustainable due to reduced annualized yield, soil health, and environmental quality. Because of similar or greater grain yields and test weights among NTWP with 50 kg N ha⁻¹ and NTWP and NTCW with other N rates, NTWP with reduced N rates may sustain spring wheat yield and grain size but not grain protein in the northern Great Plains. Full article

Durum wheat, the world’s second most cultivated wheat species, is a staple crop, critical for global food security, including in Ethiopia where it serves as a center of diversity. However, climate change and genetic erosion threaten its genetic resources, necessitating genomic studies to support conservation and breeding efforts. This study characterized genome-wide diversity, population structure (STRUCTURE, principal coordinate analysis (PCoA), neighbor-joining trees, analysis of molecular variance (AMOVA)), and selection signatures (F_ST, Hardy–Weinberg deviations) in Ethiopian durum wheat by analyzing 376 genotypes (148 accessions) using an Illumina Infinium 25K single nucleotide polymorphism (SNP) array. A set of 7842 high-quality SNPs enabled the assessments, comparing landraces with cultivars and breeding populations. Results revealed moderate genetic diversity (mean polymorphism information content (PIC) = 0.17; gene diversity = 0.20) and identified 26 loci under selection, associated with key traits like grain yield, stress tolerance, and disease resistance. AMOVA revealed 80.1% variation among accessions, with no significant differentiation by altitude, region, or spike density. Landraces formed distinct clusters, harboring unique alleles, while admixture suggested gene flow via informal seed exchange. The findings highlight Ethiopia’s rich durum wheat diversity, emphasizing landraces as reservoirs of adaptive alleles for breeding. This study provides genomic insights to guide conservation and the development of climate-resilient cultivars, supporting sustainable wheat production globally. Full article

Accurately estimating leaves’ relative chlorophyll contents (widely represented by Soil and Plant Analysis Development (SPAD) values) across growth stages is crucial for assessing crop health, particularly in regions characterized by varying sowing dates. Unlike previous studies focusing on high-resolution UAV imagery or specific growth stages, this research incorporates satellite-derived texture indices (TIs) into a SPAD value estimation model applicable across multiple growth stages (from tillering to grain-filling). Field experiments were conducted in Jiangsu Province, China, where winter wheat sowing dates varied significantly from field to field. Sentinel-2 imagery was employed to extract vegetation indices (VIs) and TIs. Following a two-step variable selection method, Random Forest (RF)-LassoCV, five machine learning algorithms were applied to develop estimation models. The newly developed model (SVR-RBF_VIs+TIs) exhibited robust estimation performance (R² = 0.8131, RMSE = 3.2333, RRMSE = 0.0710, and RPD = 2.3424) when validated against independent SPAD value datasets collected from fields with varying sowing dates. Moreover, this optimal model also exhibited a notable level of transferability at another location with different sowing times, wheat varieties, and soil types from the modeling area. In addition, this research revealed that despite the lower resolution of satellite imagery compared to UAV imagery, the incorporation of TIs significantly improved estimation accuracies compared to the sole use of VIs typical in previous studies. Full article

Foliar application of biostimulants can be a valid option to reach the goal of sustainable intensification in agriculture, especially in extensive crops such as durum wheat. However, due to the wide range of active ingredients and their mixtures available in the market, the need to select the most efficient product in a specific growing environment is of dramatic importance to achieve remarkable results in yield and grain quality. To analyze the potential of different active ingredients, a field trial was performed in two consecutive growing seasons (2023 and 2024) under Mediterranean climatic conditions. A randomized block design with three replicates was used. Durum wheat cultivar “Iride” was treated with the following five foliar biostimulants in comparison with the untreated control (T0): seaweed and plant extracts (T1); micronized vaterite (T2); culture broth of Pseudomonas protegens (T3); humic and fulvic acids (T4); organic nitrogen fertilizer (N 5%) containing glycine betaine (T5). Biostimulant treatment was applied at the end of tillering and at heading. Root length, chlorophyll content, grain yield, yield components and grain quality were measured and subjected to a one-way analysis of variance. As compared to the control, seaweed and plant extracts as well as micronized vaterite showed the best results in terms of grain yield (29% and 24% increase, respectively), root length (120% and 77% increase, respectively) and grain protein content (one percentage point increase, from approx. 12% to 13%). The results from this study can help Mediterranean farmers and researchers to develop new fertilization protocols to reach the goals of the “Farm to Fork” European strategy. Full article

In agricultural practice, in addition to determining the nitrogen (N_f) dose, it is necessary to effectively control its effect on currently grown crops. Meeting these conditions requires not only the use of phosphorus (P) and potassium (K), but also nutrients such as magnesium (Mg) and sulfur (S). This hypothesis was verified in a single-factor field experiment with winter wheat (WW) carried out in the 2015/2016, 2016/2017, and 2017/2018 growing seasons. The experiment consisted of seven variants: absolute control (AC), NP, NPK-MOP (K as Muriate of Potash), NPK-MOP+Ki (Kieserite), NPK-KK (K as Korn–Kali), NPK-KK+Ki, and NPK-KK+Ki+ES (Epsom Salt). The use of K as MOP increased grain yield (GY) by 6.3% compared to NP. In the NPK-KK variant, GY was 13% (+0.84 t ha⁻¹) higher compared to NP. Moreover, GYs in this fertilization variant (FV) were stable over the years (coefficient of variation, CV = 9.4%). In NPK-KK+Ki+ES, the yield increase was the highest and mounted to 17.2% compared to NP, but the variability over the years was also the highest (CV ≈ 20%). The amount of N in grain N (GN) increased progressively from 4% for NPK-MOP to 15% for NPK-KK and 25% for NPK-KK+Ki+ES in comparison to NP. The nitrogen harvest index was highly stable, achieving 72.6 ± 3.1%. All analyzed NUE indices showed a significant response to FVs. The PFP-N_f (partial factor productivity of N_f) indices increased on NPK-MOP by 5.8%, NPK-KK by 12.9%, and NPK-KK+Ki+ES by 17.9% compared to NP. The corresponding N_f recovery of N_f in wheat grain was 47.2%, 55.9%, and 64.4%, but its total recovery by wheat (grain + straw) was 67%, 74.5%, and 87.2%, respectively. In terms of the theoretical and practical value of the tested indexes, two indices, namely, NUP (nitrogen unit productivity) and NUA (nitrogen unit accumulation), proved to be the most useful. From the farmer’s production strategy, FV with K applied in the form of Korn–Kali proved to be the most stable option due to high and stable yield, regardless of weather conditions. The increase in the number of nutritional factors optimizing the action of nitrogen in winter wheat caused the phenomenon known as the “scissors effect”. This phenomenon manifested itself in a progressive increase in nitrogen unit productivity (NUP) combined with a regressive trend in unit nitrogen accumulation (NUA) in the grain versus the balance of soil available Mg (Mg_b). The studies clearly showed that obtaining grain that met the milling requirements was recorded only for NUA above 22 kg N t⁻¹ grain. This was possible only with the most intensive Mg treatment (NPK-KK+Ki and NPK-KK+Ki+ES). The study clearly showed that three of the six FVs fully met the three basic conditions for sustainable crop production: (i) stabilization and even an increase in grain yield; (ii) a decrease in the mass of inorganic N in the soil at harvest, potentially susceptible to leaching; and (iii) stabilization of the soil fertility of P, K, and Mg. Full article

The review summarizes the historical and current research on perennial grain breeding in Russia within the context of growing global interest in perennial crops. N.V. Tsitsin’s pioneering work in the 1930s produced the first wheat–wheatgrass amphiploids, which demonstrated the capacity to regrow after harvest and survive for 2–3 years. Subsequent research at the Main Botanical Garden in Moscow focused on characterizing Tsitsin’s material, selecting superior germplasm, and expanding genetic diversity through new cycles of hybridization and selection. This work led to the development of a new crop species, Trititrigia, and the release of cultivar ‘Pamyati Lyubimovoy’ in 2020, designed for dual-purpose production of high-quality grain and green biomass. Intermediate wheatgrass (Thinopyrum intermedium) is native to Russia, where several forage cultivars have been released and cultivated. Two large-grain cultivars (Sova and Filin) were developed from populations provided by the Land Institute and are now grown by farmers. Perennial rye was developed through interspecific crosses between Secale cereale and S. montanum, demonstrating persistence for 2–3 years with high biomass production and grain yields of 1.5–2.0 t/ha. Hybridization between Sorghum bicolor and S. halepense resulted in two released cultivars of perennial sorghum used primarily for forage production under arid conditions. Russia’s agroclimatic diversity in agricultural production systems provides significant opportunities for perennial crop development. The broader scientific and practical implications of perennial crops in Russia extend to climate-resilient, sustainable agriculture and international cooperation in this emerging field. Full article

Crop disease identification is a pivotal research area in smart agriculture, forming the foundation for disease mapping and targeted prevention strategies. Among the most prevalent global wheat diseases, powdery mildew—caused by fungal infection—poses a significant threat to crop yield and quality, making early and accurate detection crucial for effective management. In this study, we present QY-SE-MResNet34, a deep learning-based classification model that builds upon ResNet34 to perform multi-class classification of wheat leaf images and assess powdery mildew severity at the single-leaf level. The proposed methodology begins with dataset construction following the GBT 17980.22-2000 national standard for powdery mildew severity grading, resulting in a curated collection of 4248 wheat leaf images at the grain-filling stage across six severity levels. To enhance model performance, we integrated transfer learning with ResNet34, leveraging pretrained weights to improve feature extraction and accelerate convergence. Further refinements included embedding a Squeeze-and-Excitation (SE) block to strengthen feature representation while maintaining computational efficiency. The model architecture was also optimized by modifying the first convolutional layer (conv1)—replacing the original 7 × 7 kernel with a 3 × 3 kernel, adjusting the stride to 1, and setting padding to 1—to better capture fine-grained leaf textures and edge features. Subsequently, the optimal training strategy was determined through hyperparameter tuning experiments, and GrabCut-based background processing along with data augmentation were introduced to enhance model robustness. In addition, interpretability techniques such as channel masking and Grad-CAM were employed to visualize the model’s decision-making process. Experimental validation demonstrated that QY-SE-MResNet34 achieved an 89% classification accuracy, outperforming established models such as ResNet50, VGG16, and MobileNetV2 and surpassing the original ResNet34 by 11%. This study delivers a high-performance solution for single-leaf wheat powdery mildew severity assessment, offering practical value for intelligent disease monitoring and early warning systems in precision agriculture. Full article