Towards Amazon Forest Restoration: Automatic Detection of Species from UAV Imagery

Abstract

Precise assessments of forest species’ composition help analyze biodiversity patterns, estimate wood stocks, and improve carbon stock estimates. Therefore, the objective of this work was to evaluate the use of high-resolution images obtained from Unmanned Aerial Vehicle (UAV) for the identification of forest species in areas of forest regeneration in the Amazon. For this purpose, convolutional neural networks (CNN) were trained using the Keras–Tensorflow package with the faster_rcnn_inception_v2_pets model. Samples of six forest species were used to train CNN. From these, attempts were made with the number of thresholds, which is the cutoff value of the function; any value below this output is considered 0, and values above are treated as an output 1; that is, values above the value stipulated in the Threshold are considered as identified species. The results showed that the reduction in the threshold decreases the accuracy of identification, as well as the overlap of the polygons of species identification. However, in comparison with the data collected in the field, it was observed that there exists a high correlation between the trees identified by the CNN and those observed in the plots. The statistical metrics used to validate the classification results showed that CNN are able to identify species with accuracy above 90%. Based on our results, which demonstrate good accuracy and precision in the identification of species, we conclude that convolutional neural networks are an effective tool in classifying objects from UAV images.

Mapping Key Indicators of Forest Restoration in the Amazon Using a Low-Cost Drone and Artificial Intelligence

“…Like Ferreira et al [28] and Moura et al [29], this study demonstrates that RPA has high potential to map relevant species in the Amazon biome automatically. Besides mapping species, this study also showed that low-cost RPA is capable of automatic mapping and delineating individual crowns of all kinds of trees in a tropical high diverse forest.…”

“…In other studies on Cecropia, Wagner et al [68] accurately identified Cecropia hololeuca using deep learning (U-Net algorithm, which performs semantic segmentation) applied on a satellite image of the Brazilian Atlantic Forest biome. Moura et al [29] accurately mapped Cecropia using faster_R-CNN_inception_v2_pets model on an RPA image, which generates a bounding box in its results. In this work, each Cecropia crown was delineated in an instance segmentation process.…”

“…Computer vision techniques, such as deep learning [24], have high potential in improving this field of research because they have revolutionized image processing [25][26][27]. When applied to low-cost RPA images, deep learning accurately identified palm species [28], six usual tree species in the Amazon [29] and the tree species of a German forest [30]. However, the Amazon biome is a high biodiversity biome [31][32][33][34]; thus, more species identification via remote sensing will be needed in the future.…”

Remotely Sensed Tree Characterization in Urban Areas: A Review

“…These limitations, with some exceptions, such as the United States, England, and other European countries, where good-quality and open LiDAR data are more easily available [74], have been addressed by the scientific community and forest managers, who have found in UAV flights a solution to the acquisition of quality LiDAR data, even with the limitation of geographic scale, representing an advance in the use of this technology in urban forestry science [81,82]. The versatility and reasonable cost of LiDAR data acquired from UAV has allowed for a more abundant use of this data source in forestry research [61]. Although many of the studies analyzed conclude that LiDAR data work well as auxiliary information, Degerickx et al [83] and Wang et al [73] found that LIDAR data could play a central role in the characterization of urban trees, as it provides the most important information for structural, and textural features of trees.…”

“…In turn, more sophisticated algorithms, such as clustering and segmentation methods [48,56], integrated into geographical information systems (GIS), have facilitated the analysis of large-scale remotely sensed data. Computational developments during the last decades brought novel artificial intelligence (AI) techniques, which tap into the ability of computers to mimic human reasoning [57], solving complex problems such as object classification and damage assessment [58][59][60][61][62]. These advances are especially recent in the field of machine learning (ML) [63], where algorithms, such as random forest (RF) and support vector machine (SVM), and more recently, neural network-based deep learning (DL), algorithms have been developed [48,59], improving classification problems, such as individual tree characterization.…”