Plant Counting of Cotton from UAS Imagery Using Deep Learning-Based Object Detection Framework

Abstract

Assessing plant population of cotton is important to make replanting decisions in low plant density areas, prone to yielding penalties. Since the measurement of plant population in the field is labor intensive and subject to error, in this study, a new approach of image-based plant counting is proposed, using unmanned aircraft systems (UAS; DJI Mavic 2 Pro, Shenzhen, China) data. The previously developed image-based techniques required a priori information of geometry or statistical characteristics of plant canopy features, while also limiting the versatility of the methods in variable field conditions. In this regard, a deep learning-based plant counting algorithm was proposed to reduce the number of input variables, and to remove requirements for acquiring geometric or statistical information. The object detection model named You Only Look Once version 3 (YOLOv3) and photogrammetry were utilized to separate, locate, and count cotton plants in the seedling stage. The proposed algorithm was tested with four different UAS datasets, containing variability in plant size, overall illumination, and background brightness. Root mean square error (RMSE) and R2 values of the optimal plant count results ranged from 0.50 to 0.60 plants per linear meter of row (number of plants within 1 m distance along the planting row direction) and 0.96 to 0.97, respectively. The object detection algorithm, trained with variable plant size, ground wetness, and lighting conditions generally resulted in a lower detection error, unless an observable difference of developmental stages of cotton existed. The proposed plant counting algorithm performed well with 0–14 plants per linear meter of row, when cotton plants are generally separable in the seedling stage. This study is expected to provide an automated methodology for in situ evaluation of plant emergence using UAS data.

Accuracy Assessment in Convolutional Neural Network-Based Deep Learning Remote Sensing Studies—Part 2: Recommendations and Best Practices

“…a semantic segmentation). Second, mAP is generally used to indicate a composite AP, for example, typically averaged over different classes [81][82][83], but also sometimes different IoU values [84]. However, because of inconsistencies in the usage of the terms AP and mAP in the past, some sources (e.g., [85]) no longer recommend differentiating between them, and instead use the terms interchangeably.…”

“…For example, confidence intervals can be estimated for overall accuracy [97] and AUC ROC [98] or multiple model runs can be averaged to assess for variability. For example, Oh et al [82] use the mean and standard deviation of 30 replications to assess variability. It is also possible to statistically compare output.…”

Advances in Convolution Neural Networks Based Crowd Counting and Density Estimation

“…The system takes, as input, aerial images captured by an UAV equipped with RGB cameras. Sungchan et al [91] implemented an automatic cotton plant counting by adapting the Yolo3 [92] deep learning algorithm. In the same context, Kitano et al [93] used a fully convolution neural network to develop an application that captured images using a UAV and returned the number of corn plants.…”

Detection of cotton waterlogging stress based on hyperspectral images and convolutional neural network

“…These proves the possibility for multispectral images to detect waterlogging stress of cotton. RGB images are common materials for classification with deep learning [38][39][40] . This paper proved that hyperspectral images of cotton leaves with different waterlogging stress could be classified with CNN models.…”