Search

Peanut is a significant oilseed crop that is often affected by peanut southern blight, a disease that greatly reduces crop yield and quality. Therefore, accurate and timely monitoring of this disease is crucial to ensure crop safety and minimize the need for pesticides. Spectral features combined with texture features have been widely applied in plant disease monitoring. However, previous studies have mostly used original texture features, and its combination form has been rarely considered. This study presents a novel approach for monitoring peanut southern blight, integrating multiple spectral indices and textural indices (TIs). Firstly, a total of 20 vegetation indices (VIs) were extracted from the unmanned aerial vehicle multispectral images, while three TIs were constructed based on original textural features. Subsequently, Otsu-CIgreen algorithm was used to find the optimal threshold to eliminate the complex background of the image. Lastly, monitoring models for peanut southern blight were constructed using three machine learning models based on the screened VIs, VIs combined with TIs. Among these models, the K-nearest neighbor model using VIs combined with TIs demonstrates the best performance, with accuracy and F1 score on the test set reaching 91.89% and 91.39% respectively. The results indicate that the monitoring models utilizing VIs and TIs were more effective compared to models using only VIs. This approach provides valuable insights for non-destructive and accurate monitoring of peanut southern blight. [ABSTRACT FROM AUTHOR]

Accurately identifying the quantity of maize seedlings is useful in improving maize varieties with high seedling emergence rates in a breeding program. The traditional method is to calculate the number of crops manually, which is labor-intensive and time-consuming. Recently, observation methods utilizing a UAV have been widely employed to monitor crop growth due to their low cost, intuitive nature and ability to collect data without contacting the crop. However, most investigations have lacked a systematic strategy for seedling identification. Additionally, estimating the quantity of maize seedlings is challenging due to the complexity of field crop growth environments. The purpose of this research was to rapidly and automatically count maize seedlings. Three models for estimating the quantity of maize seedlings in the field were developed: corner detection model (C), linear regression model (L) and deep learning model (D). The robustness of these maize seedling counting models was validated using RGB images taken at various dates and locations. The maize seedling recognition rate of the three models were 99.78% (C), 99.9% (L) and 98.45% (D) respectively. The L model can be well adapted to different data to identify the number of maize seedlings. The results indicated that the high-throughput and fast method of calculating the number of maize seedlings is a useful tool for maize phenotyping. [ABSTRACT FROM AUTHOR]