Computers and Electronics in Agriculture最新文献

{"title":"Structural analysis and optimal design of a centrifugal side-throw organic fertiliser spreader","authors":"Chengsai Fan ,&nbsp;Xiaofeng Liu ,&nbsp;Junyang Shi ,&nbsp;Yinyan Shi ,&nbsp;Ruiyin He","doi":"10.1016/j.compag.2025.110309","DOIUrl":"10.1016/j.compag.2025.110309","url":null,"abstract":"<div><div>This study contradicts the discrepancy between limited greenhouse spaces and substantial volumes of organic fertiliser utilised. Using EDEM numerical simulation software, we analysed the contact between the centrifugal side-throw organic fertiliser-spreading disc and the organic fertiliser particles. The factors examined included fan inclination angle, disc speed, and angle of guide vanes. The coefficient of variation was used as the measure. A three-factor, three-level Box–Behnken Design (BBD) response surface test was performed to procure response surface plots, fit the data, and determine the optimal values. To validate our findings, a specific test was conducted to evaluate the efficiency of the guide vanes. The superiority of the fertiliser-spreading structure was verified by repeating it three times in field trials. Our simulation indicated the optimal values for the rotational speed of the fertiliser-spreading disc and the inclination angle of the fan blade to ensure uniform fertiliser distribution. We discovered that an excessive number of guide vanes can hinder the smooth application of organic fertilisers. In addition, the guide vanes set at wider angles demonstrated superior dispersion. Precise optimisation identified a disc rotational speed of 417 r·min⁻¹, fan inclination of 16.67°, and guide vane angle of 20°. With these settings, the ideal coefficient of variation for the lateral distribution of organic fertiliser was 15.73%, and below 16.48% in the field operation. In conclusion, we designed an adjustable centrifugal side-throw organic fertiliser-spreading disc to address the challenges of organic fertiliser distribution in greenhouses. This offers a foundational model for mechanising organic fertiliser application in agricultural facilities.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"234 ","pages":"Article 110309"},"PeriodicalIF":7.7,"publicationDate":"2025-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143724480","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Predicting the spatial distribution of reducing sugars using near-infrared hyperspectral imaging and chemometrics: a study in multiple potato genotypes","authors":"Carlos Miguel Peraza-Alemán ,&nbsp;Silvia Arazuri ,&nbsp;Carmen Jarén ,&nbsp;Jose Ignacio Ruiz de Galarreta ,&nbsp;Leire Barandalla ,&nbsp;Ainara López-Maestresalas","doi":"10.1016/j.compag.2025.110323","DOIUrl":"10.1016/j.compag.2025.110323","url":null,"abstract":"<div><div>The determination of reducing sugars in potatoes is important due to their impact on product quality during industrial processing. The significant variability of these compounds between genotypes presents a challenge to the development of accurate predictive models. This study evaluated the potential of near-infrared hyperspectral imaging (NIR-HSI) for the prediction of reducing sugars in potatoes. For this, a wide range of genotypes (n = 92) from two seasons (2020–2021) was selected. Partial Least Squares Regression (PLSR) and Support Vector Machine Regression (SVMR) methods were used to build the prediction models. Furthermore, interval PLS (iPLS), recursive weighted PLS (rPLS), Genetic Algorithm (GA) and Competitive Adaptive Reweighted Sampling (CARS) were used for relevant wavelength identification to develop less computationally complex models. The best full spectrum model (SNV-PLSR) achieved coefficient of determination and root mean square error values of 0.88 and 0.053 % and 0.86 and 0.057 %, for calibration and external validation, respectively. Variable selection algorithms successfully reduced the dimensionality of the data without compromising the performance of the models. Robust predicted models were built with only 2.65 % (CARS-PLSR) and 3.57 % (iPLS-SVMR) of the total wavelengths. Finally, a pixel-wise prediction was performed on the validation set and chemical images were built to visualise the spatial distribution of reducing sugars. This study demonstrated that NIR-HSI is a feasible technique for predicting reducing sugars in several potato genotypes.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"235 ","pages":"Article 110323"},"PeriodicalIF":7.7,"publicationDate":"2025-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143705577","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"WSG-P2PNet: A deep learning framework for counting and locating wheat spike grains in the open field environment","authors":"Qing Geng ,&nbsp;Xin Xu ,&nbsp;Xinming Ma ,&nbsp;Li Li ,&nbsp;Fan Xu ,&nbsp;Bingbo Gao ,&nbsp;Yuntao Ma ,&nbsp;Jianxi Huang ,&nbsp;Jianyu Yang ,&nbsp;Xiaochuang Yao","doi":"10.1016/j.compag.2025.110314","DOIUrl":"10.1016/j.compag.2025.110314","url":null,"abstract":"<div><div>The number of spike grains is an important parameter for wheat yield estimation. However, it is challenging to automatically and intelligently count wheat spike grains in the open field environment. In this study, a deep learning framework, called Wheat Spike Grain Point-to-Point Network (WSG-P2PNet), is proposed to count and locate the wheat spike grains in the open field environment. This framework incorporates Efficient Channel Attention (ECA) and Coordinate Attention (CA) after feature extraction and feature concatenation, respectively. These mechanisms effectively highlight the channel features and positional information of the wheat spike grains while suppressing background interference from factors such as stems, leaves and wheat ears. Additionally, standard convolutions in the regression and classification branches are replaced with Spatial and Channel reconstruction Convolutions (SCConv), further enhancing representational capabilities and improving model performance. The results demonstrate that WSG-P2PNet, using VGG19_bn as the backbone network, outperforms five other state-of-the-art methods in terms of accuracy and stability, with an MAE of 1.72 (95% CI 1.67, 1.77), an Acc of 94.93% (95% CI 94.92, 94.93), an RMSE of 2.35 (95% CI 2.26, 2.44), and an <em>R</em>² of 0.8311 (95% CI 0.8218, 0.8404). Ablation experiments illustrate the impact of SCConv, ECA, and CA on the performance of WSG-P2PNet. Notably, WSG-P2PNet still maintains high accuracy in different varieties and growth periods, demonstrating its robustness and generalizability in real-world scenarios. Preliminary experiments also evaluated the correlation between predicted spike grain numbers and wheat yield, with an average Pearson Correlation Coefficient <em>r</em> of 0.7944, indicating a strong positive statistical relationship. The proposed deep learning framework enables rapid and accurate counting and localization of wheat spike grains in the open field environment, which is of great significant for integrated wheat yield estimation.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"235 ","pages":"Article 110314"},"PeriodicalIF":7.7,"publicationDate":"2025-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143716216","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"High-fidelity 3D reconstruction of peach orchards using a 3DGS-Ag model","authors":"Yanan Chen ,&nbsp;Ke Xiao ,&nbsp;Guandong Gao ,&nbsp;Fan Zhang","doi":"10.1016/j.compag.2025.110225","DOIUrl":"10.1016/j.compag.2025.110225","url":null,"abstract":"<div><div>Accurate reconstruction of 3D orchards plays a key role in phenotyping within the field of digital agriculture. However, the model aliasing caused by occlusion presents significant challenges to high-precision 3D reconstruction during the orchard modeling process. In this paper, a 3DGS-Ag model based on improved 3D Gaussian Splatting (3DGS), is proposed to achieve high-quality reconstruction of 3D orchard scenes, taking peach orchards as an example. Datasets for three different scales of peach orchards, including multiple peach trees, a single peach tree and fruit-bearing peach trees, are created using multi-view images. In the process of adaptive density control, a dynamic opacity reset strategy is proposed to replace the reset strategy of baseline 3DGS by constructing an opacity reset function, which reduces erroneous shear during densification, achieving effective capture of scene features at different scales. In reconstructing the 3D orchard scenes, a distance-weighted filtering module is introduced, which is supervised by additional distance information to limit the representation frequency of Gaussian primitives, while integrating with the super-sampling technique to increase the sampling density of pixels. Experimental results demonstrate that the 3DGS-Ag model surpasses the 3DGS and the latest 2DGS concerning the evaluation metrics of PSNR, SSIM, and LPIPS. Specifically, it achieves improvement of 9.56% and 12.80% in PSNR, 13.67% and 12.20% in SSIM, and reduction of 21.14% and 10.75% in LPIPS, respectively. In summary, the 3DGS-Ag model proposed can exhibit higher precision in reconstructing peach orchards across multiple scales, providing valuable reference and support for advancing 3D digitization in agricultural scenes.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"234 ","pages":"Article 110225"},"PeriodicalIF":7.7,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143724104","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Reconstruction and spatial distribution analysis of maize seedlings based on multiple clustering of point clouds","authors":"Xinmin Song,&nbsp;Tao Cui,&nbsp;Dongxing Zhang,&nbsp;Li Yang,&nbsp;Xiantao He,&nbsp;Kailiang Zhang","doi":"10.1016/j.compag.2025.110196","DOIUrl":"10.1016/j.compag.2025.110196","url":null,"abstract":"<div><div>This study proposes a method for maize seedling reconstruction and spatial distribution analysis based on ground-based laser three-dimensional point cloud scanning technology. Using high-precision terrestrial laser scanning (TLS), 3D point cloud data was collected from multiple maize seedling plots, followed by detailed preprocessing and analysis using Trimble Realworks. During the data processing, a regression-based empirical formula, grounded in maize seedling growth characteristics, was proposed. This formula effectively mitigates the challenges of leaf occlusion in densely planted conditions, providing a solution for further point cloud segmentation and analysis. In terms of algorithm design, this study combines DBSCAN and K-means clustering algorithms to effectively overcome the challenges posed by the dense distribution of plants, leaf occlusion, and noise in the point cloud data. Through this multi-clustering approach, plant positions and distributions were accurately identified, row and column spacing calculations were optimized, and a missing plant detection function was implemented. Furthermore, a dynamic plant height calculation method based on ground undulation was proposed, significantly improving the accuracy of plant height measurement and addressing errors caused by terrain variations. Experimental results show that the proposed algorithm achieves high accuracy and robustness across multiple experimental plots, with a plant counting accuracy rate of 98.33%, a row and column spacing deviation rate controlled within 5%, and a plant height calculation accuracy exceeding 97%. These results demonstrate the effectiveness of this method in precise measurement and spatial distribution analysis during the maize seedling stage, providing strong support for precision agriculture. In the future, with further optimization of the technology, this method could be widely applied in agricultural automation and intelligent management.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"235 ","pages":"Article 110196"},"PeriodicalIF":7.7,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143705578","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Learning hidden relationship between environment and control variables for direct control of automated greenhouse using Transformer-based model","authors":"Junseo Lee ,&nbsp;Seongil Im ,&nbsp;Jae-Seung Jeong ,&nbsp;Taek Sung Lee ,&nbsp;Soo Hyun Park ,&nbsp;Changhwan Shin ,&nbsp;Hyunsu Ju ,&nbsp;Hyung-Jun Kim","doi":"10.1016/j.compag.2025.110335","DOIUrl":"10.1016/j.compag.2025.110335","url":null,"abstract":"<div><div>Climate change poses a significant threat to agricultural sustainability and food security. Automated greenhouse systems, which provide stable and controlled environments for crop cultivation, have emerged as a promising solution. However, traditional rule-based greenhouse control algorithms struggle to determine optimal control variables due to the complex relationships between environmental variables. In response, we propose a Transformer-based model, Trans-Farmer, which predicts the control variables by considering the complex interactions among environmental variables. Trans-Farmer leverages the attention mechanism to learn the intricate relationships among the environmental variables. The encoder-decoder structure enables the translation of the environmental variables into the corresponding control variables, analogous to language translation. Experimental results demonstrate that Trans-Farmer outperforms baseline models across all the evaluation metrics, achieving superior accuracy and predictive performance. The attention maps of the encoder visualize how Trans-Farmer comprehends the complex interactions among the environmental variables. Additionally, the compact size of Trans-Farmer is suitable for application in general greenhouses with constrained microcontroller units. This approach contributes to the development of automated greenhouse management systems and emphasizes the potential of artificial intelligence applications in agriculture.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"235 ","pages":"Article 110335"},"PeriodicalIF":7.7,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143697535","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Design and experiment of a high-speed row cleaning unit with double air spring active control system","authors":"Xiaomeng Xia ,&nbsp;Dongyan Huang ,&nbsp;Honggang Li ,&nbsp;Ruiqiang Ran ,&nbsp;Shuyan Liu ,&nbsp;Lili Fu ,&nbsp;Yongjian Cong","doi":"10.1016/j.compag.2025.110292","DOIUrl":"10.1016/j.compag.2025.110292","url":null,"abstract":"<div><div>The no-till planter needs to be equipped with a row clearer that remove plant residues from the soil surface to provide a clean seedbed for correct seeding. Consistent working depth is the key to ensuring the straw removal performance of the row cleaner. At present, applying a controlled downforce to the row cleaner is an effective method of stabilizing the working depth. However, the traditional active force control system (TACS), which can only control downforce, is ineffective in stabilising the high-speed row cleaner’s working depth as it struggles to meet the high downforce demands of high-speed row cleaner. Therefore, this study designed a high-speed row cleaning unit with double air spring active control system (DSACS). DSACS, which enabled synergistic control of stiffness and active forces, was used to realize the trade-off between stability of the row clearer’s working depth and downforce requirements at high operating speeds. The forces output from DSACS was decided by the variable universe fuzzy control algorithm, and the optimal stiffness of the DSACS at different speeds was obtained by simulation. The effectiveness of the DSACS was analyzed through simulation and field experiments. The simulation results showed that DSACS had more advantages in optimizing the dynamic performance of the row cleaning unit and reducing the active forces demands. Compared to TACS, the root mean square of impact forces decreased by 13.8 %, 7.3 %, and 12.3 %, and the root mean square of active forces decreased by 14.2 %, 9.2 %, and 12.7 % at speeds of 8 km∙h⁻¹, 10 km∙h⁻¹, and 12 km∙h⁻¹, respectively. The field experiments results showed that compared to TACS, the row cleaning unit with DSACS exhibited better straw removal performance at high operating speeds, with the reduction of 26.3 %, 25.0 % and 22.1 % in the coefficient of variation of cleaned strip width and increase of 4.8 %, 3.4 % and 5.4 % in the straw cleaning rate at the speeds of 8 km∙h⁻¹, 10 km∙h⁻¹ and 12 km∙h⁻¹, respectively.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"234 ","pages":"Article 110292"},"PeriodicalIF":7.7,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143724103","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An open-source data-driven automatic road extraction framework for diverse farmland application scenarios","authors":"Jing Shen ,&nbsp;Yawen He ,&nbsp;Jian Peng ,&nbsp;Tang Liu ,&nbsp;Chenghu Zhou","doi":"10.1016/j.compag.2025.110330","DOIUrl":"10.1016/j.compag.2025.110330","url":null,"abstract":"<div><div>The narrow contours of farmland roads, lack of clear boundary features with surrounding objects, and the complexity and variability of features limit the applicability of existing supervised extraction algorithms. Meanwhile, visual segmentation models represented by SAM (Segment Anything Model) can achieve zero-shot generalization with appropriate prompts but struggle to capture linear object effectively. This study introduces OSAM (OpenStreetMap SAM), which fine-tunes SAM using historical open-source datasets to enhance its ability to detect linear features. Then the OSAM framework dynamically generates prompts from the open geographic database OpenStreetMap to activate SAM, enabling autonomous detection of farmland roads without the need for additional manual annotations or assisted interactions. Experiments demonstrate that OSAM performs exceptionally well in scenarios with sparse farmland road distributions and delivers robust results even with limited training data. Specifically, OSAM achieves a F1 of 71.91 % and an IoU of 58.53 % when trained on the full dataset, significantly outperforming DLinkNet (IoU: 56.42 %) and SegFormer (IoU: 41.65 %). Even with only 1 % of the original training samples, OSAM maintains robust performance (F1: 62.26 %, IoU: 47.02 %), whereas supervised learning methods such as SegFormer, SIINet, and UNet suffer significant performance degradation under extreme data constraints. Furthermore, evaluations on remote sensing images with varying data distributions, spatial resolutions, and agricultural environments confirm that OSAM achieves high extraction accuracy and strong generalization ability. This framework significantly reduces reliance on large, well-balanced labeled datasets while maintaining high accuracy, making farmland road extraction more efficient and cost-effective in diverse scenarios.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"235 ","pages":"Article 110330"},"PeriodicalIF":7.7,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143705579","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Autonomous Cellular-Networked surveillance system for coconut rhinoceros beetle","authors":"Mohsen Paryavi ,&nbsp;Keith Weiser ,&nbsp;Michael Melzer ,&nbsp;Reza Ghorbani ,&nbsp;Daniel Jenkins","doi":"10.1016/j.compag.2025.110310","DOIUrl":"10.1016/j.compag.2025.110310","url":null,"abstract":"<div><div>A biological invasion of the Coconut Rhinoceros Beetle (CRB; <em>Oryctes rhinoceros</em>) to the island of Oahu was discovered in late 2013, posing a threat to palm trees on the island and potential for accidental export to other Hawaiian Islands and sub-tropical palm growing regions of California and Florida. Delineation of populations by physical trapping in remote, undeveloped areas is a critical part of the program for containment and eradication. Continuous surveillance near ports of entry is especially important to eliminate incipient populations rapidly and mitigate the risk of human-assisted transport. Traditional trap monitoring for the CRB is labor-intensive, costly, and temporally inadequate. We have developed an autonomous trap surveillance system framework using electronic sensors and front and backend remote cloud systems for monitoring the CRB trap contents. The customized surveillance system incorporates a camera and digital microphone, and communicates data through a cellular network using Category-M (CAT-M) Low-Power Wide-Area Network (LPWAN) with an integrated GNSS chip for precise geolocation of catches. Hourly monitoring data from early deployments of the system have demonstrated that adult CRB have a crepuscular behavior, with over two-thirds of catches occurring after sunset within three hours of twilight, and fewer than 1% occurring unambiguously during daylight. The system represents a significant advance for trap monitoring, and can prove valuable for identifying biological behaviors that might be exploited for more effective control.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"235 ","pages":"Article 110310"},"PeriodicalIF":7.7,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143705580","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Recent advances in pig behavior detection based on information perception technology","authors":"Jinyang Xu ,&nbsp;Yibin Ying ,&nbsp;Dihua Wu ,&nbsp;Yilei Hu ,&nbsp;Di Cui","doi":"10.1016/j.compag.2025.110327","DOIUrl":"10.1016/j.compag.2025.110327","url":null,"abstract":"<div><div>Global demand for meat is increasing with the world’s population growth, which leads to the expansion of global pig breeding. For farming enterprises, the production efficiency is important. In the process of pig breeding, pig behavior will reflect some pieces of information such as health, welfare, and growth status, which indirectly impacts the production efficiency of farming enterprises. Therefore, the detection of daily pig behaviors is essential. With the development of sensing and artificial intelligence technologies, various information perception technologies have been used in pig behavior detection. This paper provides a comprehensive review of recent advances in information perception technology for pig behavior detection. The merits and demerits of different information perception technologies for pig target perception and behavior detection were analyzed first. Then different detection systems for pig behavior were compared. Subsequently, the public datasets for pig behavior were innovatively summarized. Based on these findings, this study identifies key challenges that persist in the application of information perception technologies for pig behavior detection. These challenges include the limited data dimensionality when using a single sensing modality, the difficulty of accurately perceiving individual behavioral information in group housing conditions, the uneven research focus across different types of behaviors, the limited variety and scale of publicly available pig behavior datasets, and the heavy reliance on manual data annotation. To address these issues, future research should integrate multiple sensing modalities to enrich data quality and dimensionality, develop target extraction and behavior detection models that balance accuracy with computational complexity, broaden the scope of studied behaviors to include those previously overlooked, construct more diverse and sufficiently large datasets, and adopt semi-supervised or unsupervised strategies for data annotation. This work will facilitate large-scale commercial applications of pig behavior detection and will lay a critical foundation for welfare-oriented pig farming.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"235 ","pages":"Article 110327"},"PeriodicalIF":7.7,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143705583","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}