To enhance the accuracy of image scoring and consistency in screening for the propagation potential of succulent leaves, we propose a breeding scoring algorithm based on multi-channel feature fusion of “surface-outline-color”. This algorithm first employs an image segmentation model to automatically detect and locate areas of succulent leaves with breeding potential within the original image, completing preliminary screening. Subsequently, it extracts leaf area, contour, color, and high-level semantic features to construct a unified multidimensional fusion vector, which is then input into a scoring model to generate breeding scores. Among these, the area feature reflects developmental fullness, the contour feature indicates morphological regularity, and the color feature measures health status. Meanwhile, semantic features extracted by a pre-trained visual model supplement deeper growth patterns and semantic associations that traditional metrics struggle to capture, further enhancing the model’s discriminative power and generalization capability. Experimental results demonstrate that this method significantly improves the efficiency and accuracy of leaf screening. Compared to traditional methods, the Pearson correlation coefficient and area under the curve improved by an average of 0.093 8 and 0.065 3, respectively, across multiple succulent datasets. The mean squared logarithmic error decreased by approximately 0.012 1, demonstrating enhanced accuracy and robustness. This effectively supports the intelligent breeding of succulent plants.
[1] Zhao L, Olivier K, Chen L. An automated image segmentation, annotation, and training framework of plant leaves by joining the SAM and the YOLOv8 models [J]. Agronomy, 2025, 15(5): 1081.
[2] Altayeb J M, Abu-Taha A H, Abu-Naser S S. Deep learning-based classification of lemon plant quality: a study on identifying good and bad quality plants using CNN [J]. International Journal of Academic Information Systems Research, 2025, 3(1): 17-22.
[3] Li Y, Liu S, Wu J, et al. Multi-scale kolmogorov-arnold network (KAN)-based linear attention network: multi-scale feature fusion with KAN and deformable convolution for urban scene image semantic segmentation [J]. Remote Sensing, 2025, 17(5): 802.
[4] Tan Y L, Li Y, Jia S H, et al. Improving coniferous forests leaf area index estimation by filling the occluded point cloud from airborne laser scanning [J], Measurement, 2025, 242: 115866.
[5] Singh C, Randhawa G S, Farooque A A, et al. Evaluation of deep learning models for RGB image-based detection of potato virus y strain symptoms (O, NO, and NTN) in potato plants [J]. Smart Agricultural Technology, 2025, 10: 100755.
[6] Rai R, Bansal P. Accurate crop disease identification and classification in smart agriculture using a three-tier model and optimized fully conventional network [J]. Multimedia Tools and Applications, 2025, 84: 2539-2564.
[7] Estrada F, Gonzàlez-Meler M A, de Dias Oliveira E A, et al. Morphophysiological plant phenotyping for the development of plant breeding under drought and heat conditions: a practical approach [J]. Food and Energy Security, 2025, 14(1): e70030.
[8] Ma X, Wu Y, Lin Z, et al. TomPhenoNet: a multi-modal fusion and multi-task learning network model for monitoring growth parameters of dwarf tomatoes [J]. Computers and Electronics in Agriculture, 2025, 235: 110387.
[9] Wang H, Yan S, Wang W, et al. Cropformer: an interpretable deep learning framework for crop genomic prediction [J]. Plant Communications, 2025, 6(3): 101223.
[10] Gowthaman S, Das A. Plant leaf identification using feature fusion of wavelet scattering network and CNN with PCA classifier [J]. IEEE Access, 2025: 11594-11608.
[11] Sánchez J C M, Mesa H G A, Espinosa A T, et al. Improving wheat yield prediction through variable selection using Support Vector Regression, Random Forest, and Extreme Gradient Boosting [J]. Smart Agricultural Technology, 2025, 10: 100791.
[12] Wang M, Fan S, Li Y, et al. Robust multi-modal fusion architecture for medical data with knowledge distillation [J]. Computer Methods and Programs in Biomedicine, 2025, 260: 108568.
[13] Li X, Zhou Y, Li Y, et al. HSDT-TabNet: a dual-path deep learning model for severity grading of soybean frogeye leaf spot [J]. Agronomy, 2025, 15(7): 1530.
[14] Wang H, Ding J, Wang S, et al. Enhancing predictive accuracy for urinary tract infections post-pediatric pyeloplasty with explainable AI: an ensemble TabNet approach [J]. Scientific Reports, 2025, 15(1): 2455.
[15] Ostovar A, Davari D D, Dziku M. Determinants of design with multilayer perceptron neural networks: A comparison with logistic regression [J]. Sustainability, 2025, 17(6): 2611.
[16] Ghysels S, De Baets B, Reheul D, et al. Image-based yield prediction for tall fescue using random forests and convolutional neural networks [J]. Frontiers in Plant Science, 2025, 16: 1549099.
[17] Zhao Y, Pu L, Deng H, et al. Robust counting for multi-species plants based on Few-Shot learning [J]. Computers and Electronics in Agriculture, 2025, 229: 109745.
[18] Lee Y, Jeong J. TSMixer-and transfer learning-based highly reliable prediction with shortterm time series data in small-scale solar power generation systems [J]. Energies, 2025, 18(4): 765.
[19] Miftahushudur T, Sahin H M, Grieve B, et al. A survey of methods for addressing imbalance data problems in agriculture applications [J]. Remote Sensing, 2025, 17(3): 454.
[20] Pan Z, Lu Z, Li S, et al. Seasonal variation in root morphological traits and non-structural carbohydrates of pinus yunnanensis seedlings across different seedling orders [J]. Plants, 2025, 14(5): 825.
[21] Eckersley J, Moore C E, Thompson S E, et al. Separating leaf area index from plant area index using semi-supervised classification of digital hemispheric canopy photographs: a case study of dryland vegetation [J]. Agricultural and Forest Meteorology, 2025, 363: 110395.
[22] Khanam R, Hussain M. Yolov11: an overview of the key architectural enhancements [DB/OL]. (2024-10-23) [2025-12-20]. https://arxiv.org/abs/2410.17725.
[23] Zhang W, Shen L, Foo C S. Source-free domain adaptation guided by vision and visionlanguage pre-training [J]. International Journal of Computer Vision, 2025, 133(2): 844-866.
[24] Zheng Z, Liu K, Zhou Y, et al. Response to letter to the Editor from Y. Takefuji on “Beyond principal component analysis: enhancing feature reduction in electronic noses through robust statistical methods” [J]. Trends in Food Science & Technology, 2025, 157: 104918.
[25] Boufssasse A, Joudar N E, Ettaouil M. Novel dropout approach for mitigating oversmoothing in graph neural networks [J]. Applied Intelligence, 2025, 55(5): 1-14.
[26] Reyad M, Sarhan A M, Arafa M. A modified Adam algorithm for deep neural network optimization [J]. Neural Computing and Applications, 2023, 35(23): 17095-17112.
[27] 江会权. 基于改进的MobileNetV3多肉植物图像分类识别[J]. 农业技术与装备, 2024(5): 9-11,14. Jiang H Q. Classification and recognition of succulent plant images based on improved MobileNetV3[J]. Agricultural technology & Equipment, 2024(5): 9-11,14. (in Chinese)
[28] Arik S Ö, Pfister T. Tabnet: attentive interpretable tabular learning [C]//AAAI Conference on Artificial Intelligence, 2021: 6679-6687.
[29] Song W, Shi C, Xiao Z, et al. Autoint: automatic feature interaction learning via self-attentive neural networks [C]//28th ACM International Conference on Information and Knowledge Management, 2019: 1161-1170.
[30] Liu Z, Wang Y, Vaidya S, et al. Kan: Kolmogorov-Arnold networks [DB/OL]. (2025-02-09) [2025-12-20]. https://arxiv.org/abs/2404.19756.
[31] Chen T Q, Guestrin C. XGBoost: a scalable tree boosting system [C]//22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2026: 785-794.
[32] Ke G, Meng Q, Finley T, et al. Lightgbm: a highly efficient gradient boosting decision tree [C]//31st Conference on Neural Information Processing Systems (NIPS), 2017: 1-9.
[33] He K, Zhang X, Ren S, et al. Deep residual learning for image recognition [C]//IEEE Conference on Computer Vision and Pattern Recognition, 2016: 770-778.
[34] Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition [DB/OL]. (2015-04-10) [2025-12-20]. https://arxiv.org/abs/1409.1556.
[35] Dosovitskiy A. An image is worth 16x16 words: transformers for image recognition at scale [DB/OL]. (2021-06-03) [2025-12-10]. https://arxiv.org/abs/2010.11929.
