Automatic estimation of rice grain number based on a convolutional neural network.

The grain number on the rice panicle, which directly determines the rice yield, is a very important agronomic trait in rice breeding and yield-related research. However, manual counting of grain number per rice panicle is time-consuming, error-prone, and laborious. In this study, a novel prototype, dubbed the "GN-System," was developed for the automatic calculation of grain number per rice panicle based on a deep convolutional neural network. First, a whole panicle grain detection (WPGD) model was established using the Cascade R-CNN method embedded with the feature pyramid network for grain recognition and location. Then, a GN-System integrated with the WPGD model was developed to automatically calculate grain number per rice panicle. The performance of the GN-System was evaluated through estimated stability and accuracy. One hundred twenty-four panicle samples were tested to evaluate the estimated stability of the GN-System. The results showed that the coefficient of determination (R2) was 0.810, the mean absolute percentage error was 8.44%, and the root mean square error was 16.73. Also, another 12 panicle samples were tested to further evaluate the estimated accuracy of the GN-System. The results revealed that the mean accuracy of the GN-System reached 90.6%. The GN-System, which can quickly and accurately predict the grain number per rice panicle, can provide an effective, convenient, and low-cost tool for yield evaluation, crop breeding, and genetic research. It also has great potential in assisting phenotypic research.

PDC: Pearl Detection with a Counter Based on Deep Learning.

Pearl detection with a counter (PDC) in a noncontact and high-precision manner is a challenging task in the area of commercial production. Additionally, sea pearls are considered to be quite valuable, so the traditional manual counting methods are not satisfactory, as touching may cause damage to the pearls. In this paper, we conduct a comprehensive study on nine object-detection models, and the key metrics of these models are evaluated. The results indicate that using Faster R-CNN with ResNet152, which was pretrained on the pearl dataset, mAP@0.5IoU = 100% and mAP@0.75IoU = 98.83% are achieved for pearl recognition, requiring only 15.8 ms inference time with a counter after the first loading of the model. Finally, the superiority of the proposed algorithm of Faster R-CNN ResNet152 with a counter is verified through a comparison with eight other sophisticated object detectors with a counter. The experimental results on the self-made pearl image dataset show that the total loss decreased to 0.00044. Meanwhile, the classification loss and the localization loss of the model gradually decreased to less than 0.00019 and 0.00031, respectively. The robust performance of the proposed method across the pearl dataset indicates that Faster R-CNN ResNet152 with a counter is promising for natural light or artificial light peal detection and accurate counting.

Automated Counting Grains on the Rice Panicle Based on Deep Learning Method.

Grain number per rice panicle, which directly determines grain yield, is an important agronomic trait for rice breeding and yield-related research. However, manually counting grains of rice per panicle is time-consuming, laborious, and error-prone. In this research, a grain detection model was proposed to automatically recognize and count grains on primary branches of a rice panicle. The model used image analysis based on deep learning convolutional neural network (CNN), by integrating the feature pyramid network (FPN) into the faster R-CNN network. The performance of the grain detection model was compared to that of the original faster R-CNN model and the SSD model, and it was found that the grain detection model was more reliable and accurate. The accuracy of the grain detection model was not affected by the lighting condition in which images of rice primary branches were taken. The model worked well for all rice branches with various numbers of grains. Through applying the grain detection model to images of fresh and dry branches, it was found that the model performance was not affected by the grain moisture conditions. The overall accuracy of the grain detection model was 99.4%. Results demonstrated that the model was accurate, reliable, and suitable for detecting grains of rice panicles with various conditions.

A Plant Leaf Geometric Parameter Measurement System Based on the Android Platform.

Automatic and efficient plant leaf geometry parameter measurement offers useful information for plant management. The objective of this study was to develop an efficient and effective leaf geometry parameter measurement system based on the Android phone platform. The Android mobile phone was used to process and measure geometric parameters of the leaf, such as length, width, perimeter, and area. First, initial leaf images were pre-processed by some image algorithms, then distortion calibration was proposed to eliminate image distortion. Next, a method for calculating leaf parameters by using the positive circumscribed rectangle of the leaf as a reference object was proposed to improve the measurement accuracy. The results demonstrated that the test distances from 235 to 260 mm and angles from 0 to 45 degrees had little influence on the leafs' geometric parameters. Both lab and outdoor measurements of leaf parameters showed that the developed method and the standard method were highly correlated. In addition, for the same leaf, the results of different mobile phone measurements were not significantly different. The leaf geometry parameter measurement system based on the Android phone platform used for this study could produce high accuracy measurements for leaf geometry parameters.

Automatic Diagnosis of Rice Diseases Using Deep Learning.

Rice disease has serious negative effects on crop yield, and the correct diagnosis of rice diseases is the key to avoid these effects. However, the existing disease diagnosis methods for rice are neither accurate nor efficient, and special equipment is often required. In this study, an automatic diagnosis method was developed and implemented in a smartphone app. The method was developed using deep learning based on a large dataset that contained 33,026 images of six types of rice diseases: leaf blast, false smut, neck blast, sheath blight, bacterial stripe disease, and brown spot. The core of the method was the Ensemble Model in which submodels were integrated. Finally, the Ensemble Model was validated using a separate set of images. Results showed that the three best submodels were DenseNet-121, SE-ResNet-50, and ResNeSt-50, in terms of several attributes, such as, learning rate, precision, recall, and disease recognition accuracy. Therefore, these three submodels were selected and integrated in the Ensemble Model. The Ensemble Model minimized confusion among the different types of disease, reducing misdiagnosis of the disease. Using the Ensemble Model to diagnose six types of rice diseases, an overall accuracy of 91% was achieved, which is considered to be reasonably good, considering the appearance similarities in some types of rice disease. The smartphone app allowed the client to use the Ensemble Model on the web server through a network, which was convenient and efficient for the field diagnosis of rice leaf blast, false smut, neck blast, sheath blight, bacterial stripe disease, and brown spot.

Simulation analysis of fertilizer discharge process using the Discrete Element Method (DEM).

Fertilizer discharge process is a critical part of fertilizer application, as it affects the fertilizer discharge rate and uniformity of fertilizer application. In this study, a spiral grooved-wheel fertilizer discharge device was designed to replace the conventional straight grooved-wheel. Comparisons of the fertilizer discharge performance of the two grooved-wheel types were performed through tests and simulations using the discrete element method (DEM). The discharge performance of the two discharge devices was assessed by measuring the discharge mass rate, discharge uniformity, and the falling velocity of the fertilizer particles. Results showed that under similar conditions, the fertilizer discharge mass rate of the spiral grooved-wheel was higher than that of the straight grooved-wheel. The fertilizer discharge uniformity of the spiral grooved-wheel was much better than that of the straight grooved-wheel. The average falling velocity of fertilizer particles through the discharge spout was higher under the spiral grooved-wheel. The relative errors between the test and simulation results for the discharge mass rates, discharge uniformity, and particle falling velocities of the spiral grooved-wheel were all less than 10%. The developed spiral grooved-wheel exhibited a better performance than the conventional straight grooved-wheel, in all the aspects examined. The results serve as a theoretical basis for guiding the design of high-performance fertilizer applicators.