Oil Palm Detection via Deep Transfer Learning

Abstract

This article presents an intelligent system using deep learning algorithms and the transfer learning approach to detect oil palm units in multispectral photographs taken with unmanned aerial vehicles. Two main contributions come from this piece of research. First, a dataset for oil palm units detection is carefully produced and made available online. Although being tailored to the palm detection problem, the latter has general validity and can be used for any classification application. Second, we designed and evaluated a state-of-the-art detection system, which uses a convolutional neural network to extract meaningful features, and a classifier trained with the images from the proposed dataset. Results show outstanding effectiveness with an accuracy peak of 99.5% and a precision of 99.8%. Using different images for validation taken from different altitudes the model reached an accuracy of 97.5% and a precision of 98.3%. Hence, the proposed approach is highly applicable in the field of precision agriculture.

Mapeamento sistemático de detecção e contagem de plantas em imagens agrícolas usando aprendizagem de máquina – proposta de modelagem para desenvolvimento de sistema

“…Tem-se alguns destaques com Randon Forest, Neupane et al (2019), que realizaram a contagem de pés de banana, e Bonet et al (2020), que usaram SVM com o intuito de realizar a contagem de palmeiras de extração de óleo vegetal.…”

“…,Neupane et al (2019),Selim et al (2019),Barbedo et al (2020),Bonet et al (2020),Kalantar et al (2020), Valente et al (2020), Xu et al (2020), e Zhang et al (2020). Já o cálculo do F1-score foi utilizado por Csillik et al (2018), Barbedo et al (2020) e Xu et al (2020).…”

Deep Convolutional Neural Network for Large-Scale Date Palm Tree Mapping from UAV-Based Images

“…The mapping and detection of individual tree crowns, tree/plant/vegetation species, crops, and wetlands from UAV-based images are achieved by diverse CNN architectures, which are used to perform different tasks, including path-based classification [78][79][80][81][82][83][84][85][86][87], object detection [88][89][90][91][92][93][94][95][96][97], and semantic segmentation [98][99][100][101][102][103][104][105][106][107]. Recently, semantic segmentation, a commonly used term in computer vision where each pixel within the input imagery is assigned to a particular class, has been a widely used technique in diverse earth-related applications [108].…”

Self-supervised Approach for Urban Tree Recognition on Aerial Images

“…The models could act as a feature extractor or sometimes just fine-tuned to perform the same task on a different data set. This technique has been used in the area of remote sensing for tasks like determining a specific plant species [7]. The pre-built neural network model from DeepForest can be used to learn new tree features and image backgrounds by leveraging information from the existing model weights based on data from a diverse set of forests.…”