Soybean Yield Preharvest Prediction Based on Bean Pods and Leaves Image Recognition Using Deep Learning Neural Network Combined With GRNN

Abstract

Soybean yield is a highly complex trait determined by multiple factors such as genotype, environment, and their interactions. The earlier the prediction during the growing season the better. Accurate soybean yield prediction is important for germplasm innovation and planting environment factor improvement. But until now, soybean yield has been determined by weight measurement manually after soybean plant harvest which is time-consuming, has high cost and low precision. This paper proposed a soybean yield in-field prediction method based on bean pods and leaves image recognition using a deep learning algorithm combined with a generalized regression neural network (GRNN). A faster region-convolutional neural network (Faster R-CNN), feature pyramid network (FPN), single shot multibox detector (SSD), and You Only Look Once (YOLOv3) were employed for bean pods recognition in which recognition precision and speed were 86.2, 89.8, 80.1, 87.4%, and 13 frames per second (FPS), 7 FPS, 24 FPS, and 39 FPS, respectively. Therefore, YOLOv3 was selected considering both recognition precision and speed. For enhancing detection performance, YOLOv3 was improved by changing IoU loss function, using the anchor frame clustering algorithm, and utilizing the partial neural network structure with which recognition precision increased to 90.3%. In order to improve soybean yield prediction precision, leaves were identified and counted, moreover, pods were further classified as single, double, treble, four, and five seeds types by improved YOLOv3 because each type seed weight varies. In addition, soybean seed number prediction models of each soybean planter were built using PLSR, BP, and GRNN with the input of different type pod numbers and leaf numbers with which prediction results were 96.24, 96.97, and 97.5%, respectively. Finally, the soybean yield of each planter was obtained by accumulating the weight of all soybean pod types and the average accuracy was up to 97.43%. The results show that it is feasible to predict the soybean yield of plants in situ with high precision by fusing the number of leaves and different type soybean pods recognized by a deep neural network combined with GRNN which can speed up germplasm innovation and planting environmental factor optimization.

Image-based phenotyping of seed architectural traits and prediction of seed weight using machine learning models in soybean

“…In that study, apart from architectural traits, seed coat color was also documented for diversity characterization of germplasm according to UPOV classification. Recently, several pieces of literature reported the variability of seed traits captured using imaging sensors in soybean ( Yuan et al., 2019 ; Baek et al., 2020 ; Yang et al., 2021 ; Lu et al., 2022 ). However, the present work demonstrated the application of imaging sensors to predict SW from SAT, which is an important yield-contributing trait to identify superior donors for breeding better soybean crop plants.…”

Recognition of soybean pods and yield prediction based on improved deep learning model

“…Soybean is not only one of the five major crops in the world, but also an essential high protein grain and oil crop (Yu et al, 2022). As a leaf homologous organ (Lu et al, 2022), soybean pods are an important factor in determining grain yield and quality. Therefore, it is necessary to detect the pod's quality of soybean plants in different growth stages and analyze the phenotypic characters of different varieties of pods.…”

“…Ning et al (2021) proposed a phenotypic information extraction method for the soybean plant based on IM-SSD+ACO algorithm, which realized the extraction of soybean phenotypic traits including the number of pods, plant height, number of branches, main stem and plant type, effectively. Lu et al (2022) proposed a method based on the YOLOv3 algorithm to predict soybean yield according to the number of pods and leaves. Zhang et al (2021) constructed a soybean yield prediction model based on skew parameters using the color of soybean canopy leaves at different growth stages as input values.…”

Artificial Neural Network for Discrimination and Classification of Tropical Soybean Genotypes of Different Relative Maturity Groups

“…Moreover, different from parametric statistical approaches, they capture the complex characteristics of a dataset in addition to being slightly susceptible to noise and outliers and being suitable for nonlinearly separable problems common to agricultural experimentation ( Kavzoglu and Mather, 2003 ; Sudheer et al, 2003 ; Haykin, 2008 ). Currently, at the experimental level, ANN models have been used in the prediction of genetic values ( Soares et al, 2015 ), adaptability and stability ( do Carmo Oda et al, 2019 ), phenotyping ( Sá, 2018 ), yield estimates ( Lu et al, 2022 ), genetic diversity ( Rahimi et al, 2019 ; Taratuhin et al, 2020 ), disease detection, and classification ( Hang et al, 2019 ; Trivedi et al, 2021 ). Moreover, they have demonstrated that the efficiency in the breeding stages can be increased, which can reduce the time and cost of obtaining high-performance cultivars.…”