Aiming at precise evaluation of the performance of soybean seed metering devices, a photoelectric sensor-based real-time monitoring system was designed. The proposed system mainly included a photoelectric sensor module for seeding signal collecting, Hall sensors speeding module, microcontroller unit (MCU), light and sound alarm module, human–machine interface (HMI), and other parts. The indexes of miss, multiples, flow rate, and application rate were estimated on the basis of seeder speed, seed metering disk rotation rate, photoelectric sensor signals, and clock signals. These real-time statistics of the seeding process were recorded by seeding management system. The laboratory results showed that the detection errors of seeding quantity of both big- and small-diameter soybeans were less than 2.0%. Miss and multiples index estimated by this system were 0.4% and 0.5% than that of seeding image monitoring platform (SIMP), respectively. In field tests, miss and multiples index can be used to evaluate the performance of seed metering device, and big-diameter seeds can be detected more precisely than small ones by these photoelectric sensors. This system can provide support for evaluation of working performance of seed metering devices and have a positive effect on seeding monitoring technology.

The rapid and accurate detection of weeds is the prerequisite and foundation for precision weeding, automation, and intelligent field operations. Due to the wide variety of weeds in the field and their significant morphological differences, most existing detection methods can only recognize major crops and weeds, with a pressing need to enhance accuracy. This study introduces a novel weed detection approach that integrates the GFPN (Green Feature Pyramid Network), Slide Loss, and multi-SEAM (Spatial and Enhancement Attention Modules) to enhance accuracy and improve efficiency. This approach recognizes crop seedlings utilizing an improved YOLO v8 algorithm, followed by the reverse detection of weeds through graphics processing technology. The experimental results demonstrated that the improved YOLO v8 model achieved remarkable performance, with an accuracy of 92.9%, a recall rate of 87.0%, and an F1 score of 90%. The detection speed was approximately 22.47 ms per image. And when shooting from a height ranging from 80 cm to 100 cm in the field test, the crop detection effect was the best. This reverse weed detection method addresses the challenges posed by weed diversity and complexities in image recognition modeling, thereby contributing to the enhancement of automated and intelligent weeding efficiency and quality. It also provides valuable technical support for precision weeding in farmland operations.

Predicting the excretion of feces, urine and nitrogen using support vector regression: A case study with Holstein dry cows

The massive quantities of genetic data generated by high-throughput sequencing pose challenges to data storage, transmission and analyses. These problems are effectively solved through data compression, in which the size of data storage is reduced and the speed of data transmission is improved. Several options are available for compressing and storing genetic data. However, most of these options either do not provide sufficient compression rates or require a considerable length of time for decompression and loading. Here, we propose TRCMGene, a lossless genetic data compression method that uses a referential compression scheme. The novel concept of two-step compression method, which builds an index structure using K-means and k-nearest neighbours, is introduced to TRCMGene. Evaluation with several real datasets revealed that the compression factor of TRCMGene ranges from 9 to 21. TRCMGene presents a good balance between compression factor and reading time. On average, the reading time of compressed data is 60% of that of uncompressed data. Thus, TRCMGene not only saves disc space but also saves file access time and speeds up data loading. These effects collectively improve genetic data storage and transmission in the current hardware environment and render system upgrades unnecessary. TRCMGene, user manual and demos could be accessed freely from https://github.com/tangyou79/TRCM. The data mentioned in this manuscript could be downloaded from: https://github.com/tangyou79/TRCM/wiki.

: To improve the efficiency of rhizobial inoculation operations and address the current situation of rhizobial inoculation being labor-intensive, a spraying and control system for variable rate application (VRA) of liquid rhizobia was developed. The system can accurately regulate the spraying rate of liquid rhizobia according to the desired rate of rhizobia by incremental Proportional Integral Derivative (PID) closed-loop control algorithm, making it particularly suitable for precise spraying with small amounts. In the laboratory experiment of precise flow control, variation coefficient of spraying nozzle from 0.10 MP to 0.20 MP pressure were no more than 1.2%. A clear linear relationship was observed between the spraying rate and pressure. The maximum response time was 1.93 seconds, while the average response time was 1.62 seconds when controlling the spraying system for small amounts. Field trial was conducted in Heilongjiang Agricultural Xianghe Farm of China. Average seeds number and 100-seed weight of soybean plant with rhizobia inoculation treatment were better than control group, and average yield of spraying liquid rhizobia agent with full base fertilizer increased by 232.5 kg/hm2, representing a significant 7.4% improvement over the control group. The results of these trails clearly showed the positive effect of the spraying and control system for liquid rhizobial inoculation, which has superior to existing inoculation manners, and provides an important technical support for the widespread adoption of rhizobia technology.

Maintaining optimal humidity in livestock buildings during winter is a major challenge in cold climate regions due to the conflict between moisture-removing ventilation and the need for heat preservation. To address this issue, a novel condensation dehumidification system is proposed that utilizes the natural low temperature of cold winters. An integrated energy consumption model, coupling moisture and thermal balances, was developed to evaluate room temperature drop, dehumidification rate (DR), and the internal circulation coefficient of performance (IC-COP). The model was calibrated and validated with experimental data comprising over 150 operational cycles under varied operation conditions, including initial temperature differences (ranging from −20 to −5 °C), air flow rates (0.6–1.5 m/s), refrigerant flow rates (3–7 L/min), and high-humidity conditions (>90% RH). Correlation analysis showed that higher indoor humidity improved both DR and IC-COP. Four machine learning models—Extreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LightGBM), Random Forest (RF), and Multilayer Perceptron (MLP)—were developed and compared with a stacking ensemble learning model. Results demonstrated that the stacking model achieved superior prediction accuracy, with the best R² reaching 0.908, significantly outperforming individual models. This work provides an energy-saving dehumidification solution for enclosed livestock housing and a case study on the application of machine learning for energy performance prediction and optimization in agricultural environmental control.