A novel semi-supervised framework for UAV based crop/weed classification

Abstract

Excessive use of agrochemicals for weed controlling infestation has serious agronomic and environmental repercussions associated. An appropriate amount of pesticide/ chemicals is essential for achieving the desired smart farming and precision agriculture (PA). In this regard, targeted weed control will be a critical component significantly helping in achieving the goal. A prerequisite for such control is a robust classification system that could accurately identify weed crops in a field. In this regard, Unmanned Aerial Vehicles (UAVs) can acquire high-resolution images providing detailed information for the distribution of weeds and offers a cost-efficient solution. Most of the established classification systems deploying UAV imagery are supervised, relying on image labels. However, this is a time-consuming and tedious task. In this study, the development of an optimized semi-supervised learning approach is proposed, offering a semi-supervised generative adversarial network for crops and weeds classification at early growth stage. The proposed algorithm consists of a generator that provides extra training data for the discriminator, which distinguishes weeds and crops using a small number of image labels. The proposed system was evaluated extensively on the Red Green Blue (RGB) images obtained by a quadcopter in two different croplands (pea and strawberry). The method achieved an average accuracy of 90% when 80% of training data was unlabeled. The proposed system was compared with several standards supervised learning classifiers and the results demonstrated that this technique could be applied for challenging tasks of crops and weeds classification, mainly when the labeled samples are small at less training time.

Review of Current Robotic Approaches for Precision Weed Management

“…Shahbaz et al [70] proposed a semi-supervised generative adversarial network (SGAN) for crops and weeds classification in early stages of growth, and achieved an average accuracy of 90% when 80% of the training data was unlabelled. Jiang et al [71] proposed a CNN feature-based graph conventional network and combined the features of unlabelled vertices with nearby labelled ones; thus, the problem of weed and crop recognition was transferred to semi-supervised learning on a graph to reduce manual effort.…”

Generative Adversarial Networks for Image Augmentation in Agriculture: A Systematic Review

“…Evaluated on the weedNet (Sa et al, 2017), a multispec- tral (RGB+NIR) dataset acquired by a UAV from sugar beet fields, the GAN model achieved F1 scores of about 0.85 by using two-channel (Red+NIR) images with 50% labeled data (Kerdegari et al, 2019). Khan et al (2021) employed SS-GAN for classifying crops and weeds in the RGB images acquired by UAVs from pea and strawberry fields at early crop growth stages. The SS-GAN achieved an overall classification accuracy of about 90% when only 20% labeled samples were used for training, which compared favorably to the results obtained by conventional supervised classifiers (e.g., K-nearest neighbor, SVM and CNNs).…”

Unmanned aerial vehicles for surveillance and control of vectors of malaria and other vector-borne diseases

“…Consequently, it is essential for malaria researchers and control programmes to focus on novel technologies that aid the surveillance of vectors and the delivery of control agents, with Unmanned Aerial Vehicles (UAVs) being one of the promising possible additions to the toolkit [4]. The use of UAVs has seen a considerable expansion from limited military use to their being utilized in a range of scientific and industrial applications, including agricultural remote sensing [5][6][7], response to and prevention of pest outbreaks [8,9], zoonosis control [10], humanitarian emergency response [11,12], public health [13] and species monitoring for conservation [14,15]. The idea of using UAVs in malaria control has been postulated for many years [4,16].…”