Short-range photogrammetry in forestry
The use of remote sensing data in obtaining information about forests has a long, almost 100-year history. Remote sensing satellite and aerial data have a number of applications and in many countries support the management and protection of natural resources on a daily basis. Short-range photogrammetry has long been of interest to foresters, and its significant development in recent years has initiated attempts to implement such solutions into practice.
There are more and more tools on the market for remote imaging (in the form of images or point clouds) of trees and forest stands from the ground. Due to the miniaturization of devices, interest in their practical use is growing. There are a number of scientific publications that point to the potential of using short-range remote sensing data for precise tree measurements. Nevertheless, the main barrier to their use in practice is the lack of highly automated data processing solutions that, on the one hand, provide reliable information about the measured object, and on the other hand, are tailored to the needs of specialists.
Precision forestry in practice
The concept of precision forestry will be applied in order to provide practical solutions for the application of short-range photogrammetry. Precision forestry means solutions related to the detection and characterization of individual trees using remote sensing data. The result of the work of the precision forestry team will be an AI solution that allows the characteristics and quality of standing or lying individual trees to be measured. Individual tree information integrated for any area (e.g. sample plot, forest stand, park) can be used by various stakeholders. The aim of the team’s work will be to transfer the results of basic research to practical use in various fields related to environmental protection and management.
Identification of tree species and measurement of biometric characteristics
The problem of recognizing tree species has been an area of interest since the beginning of the use of remote sensing in the analysis of the forest environment. In this regard, works particularly related to aerial and satellite data acquisition systems have a long history of publication. Aerial imaging allows collection of quite precise information about the upper layer of the forest, without much detail of trees growing in the lower layers. Tree imaging from the ground level allows for precise measurements of even the youngest trees growing in the lower layers of the forest stand.
Specific expectations of practitioners in the context of measuring the characteristics of individual trees concern the measurement of: volume, tree species, diameter at chest height (thickness at 1.3 m), tree height, size class and quality, determined on the basis of defects visible on the trunk, health based on i.a. leaf color and density. The aim of the team’s work will be to develop methods that will allow for practical application and thus replacing a large part of labor-intensive field measurements used in forestry. We will look for solutions that allow easy integration of the results of automatic measurements with existing systems used in forestry or inventory of urban green areas.
Recognition of tree size and quality
The size of the trees and whether there are defects on the side of the trunk affect the economic value of individual trees. However, there may also be various parasites or effects of various biotic and abiotic factors on the sides of the trunks, which in turn tell us about the current or future health status of the trees. Short-range remote sensing provides data that is likely to help visualize various artifacts in trees. The use of artificial intelligence algorithms can further increase the probability of detecting these artifacts. Another goal of the team will be to use artificial intelligence to recognize the size and quality of standing and fallen trees in order to estimate their value.