DE102022212642A1 - Method and system for determining tree diameters at breast height - Google Patents

Abstract

The invention relates to a system (1000) and a method (2000) for determining tree diameters at chest height, which comprises the steps: capturing aerial images of an area (300) with a bare tree population (350) in sunshine using an aircraft with a camera ( 1150);Evaluating the aerial images comprising the sub-steps of determining a 3D point cloud of the area (300) using a 3D modeling unit (1300) by using a structure-from-motion method;Normalizing the 3D points with respect to a surface relief of the area (300) ;Remove the tree components and create a relief-normalized orthomosaic of the ground-level 3D points of the area (300); Classifying the image points of the normalized orthomosaic to determine the cast shadows with an artificial neural network trained via deep learning; Determining a shadow length for a tree diameter at chest height by using the position of the sun when capturing the aerial images (400) to determine the length of a vertical object at chest height corresponding to a vertical object determining the shadow length corresponding to the flat plane, and determining the diameter by determining the width of the determined cast shadows at a distance of the determined shadow length from the tree foot position that can be assigned to the shadow; and outputting the determined tree diameters at breast height using an output device (1900).

Description

The invention relates to a system and a method for determining tree diameters at breast height.

In forestry there has always been a need to carry out inventory of the tree population. During such a forest inventory, the tree diameter at breast height (BDH) of the trees is often determined. The breast height here is defined as a height of 1.30 m above the base of the tree. In the current state of the art, it is common practice to measure this circumference for a sample of individual trees in the stand using a caliper or tape measure. To do this, every tree in the area that is part of the sample must be approached. Since manual measurement is very time-consuming, only in a few exceptions is a crop completely measured in a so-called full clamping. In practice, the data is often only roughly estimated.

For the measurement, the measuring height on the tree trunk must also be determined. Although the DBH is defined at 1.3 m above the base of the tree, in practice it is usually only estimated roughly and visually. When measuring with a clip, at least two diameters must be measured and averaged (cross clip). When measuring with a tape measure, errors can easily occur if the tape measure is twisted or crooked. Furthermore, the measured values must be documented, which is often done by hand on forms. After the measurement, the digitalization of the measured values creates additional effort before they can be evaluated. There are various developments on the market designed to support this process. The starting points for new developments here are simplifying data storage through digital documentation of the measurement results.

Terrestrial, digital methods replace manual measurements with optical sensors such as LiDAR or stereo cameras etc., which can be used to collect the data. An inspection of the existing area is also necessary here, but this can be done more quickly as a single approach to a tree is not necessary, but can be done from close up. Due to the limited area coverage, data collection is usually carried out on a sample basis. Detection hardware used must be moved within the trees as the detection ranges are limited. Complete recordings of forest areas are therefore difficult.

In addition to terrestrial methods, there are remote sensing methods that are used to indirectly estimate the DBH using aerial images taken, for example, by drones, airplanes or satellites. The diameters are not measured directly here, but are derived from other tree parameters using allometric functions. Most often, tree canopy and tree height are used as input data. However, since these are model calculations, they are usually subject to significant errors. These errors are put into perspective somewhat for large-scale recording, as in this case they are usually used on a rough scale (whole company or region).

Particularly important for such an approach are regionalized models based on true environmental data (ground truth data), but these are only available to a limited extent. Multicopter drones can cover a maximum area of around 50 ha in one flight, with fixed-wing aircraft reaching up to 250 ha. When capturing aerial images from an airplane, there are virtually no relevant restrictions, but the costs for the capture are increased. Optical satellite data is limited by cloud cover. Compared to drone data, aerial data from aircraft and satellite data have lower geometric resolutions. High-resolution satellite data usually requires a fee. In order to ensure data collection at the individual tree level, resolutions of at least 30-50 cm are necessary. In this resolution range, satellite data is extremely expensive.

The invention is therefore based on the object of making it easier and more reliable to record tree diameters at breast height, especially for large forest areas and particularly preferably for individual trees.

The object is achieved according to the invention by a method with the features of claim 1 and a system with the features of claim 7. Advantageous refinements result from the subclaims.

The invention is based on the idea of flying over a forest area with an aircraft, in particular an unmanned aircraft, for example a drone, and capturing aerial images while doing so. The aerial images are captured in a time of year when the trees are leafless. Furthermore, the aerial images are captured at a time when there is direct sunlight, ie there is no obscuration by clouds. As a result, the trees cast shadows that are directly correlated with the tree trunk diameters. Based on the information captured in the aerial photographs, an evaluation is carried out to identify the cast shadows and then at a distance of one The base of the trees are measured in terms of their width, whereby the distance corresponds to a length of shadow cast by a 1.30 m high, vertical object under the corresponding sun position conditions at which the aerial images were captured.

Evaluation is carried out in particular by first generating a 3D point cloud of the area, in particular a forest area, using a structure-from-motion algorithm. This 3D point cloud represents both the relief of the area and the objects located on it, especially the trees growing on it. The points of the three-dimensional point cloud are still assigned their optical properties, i.e. H. their color and brightness, etc. First, the 3D point cloud is preferably examined with regard to unevenness in the terrain, i.e. H. of the relief, normalized this means that all height information, for example z coordinates of the remaining normalized 3D point cloud, refer to heights compared to a subsurface that is assumed to be level. 3D points that are assigned a z coordinate close to zero after normalization are therefore points close to the ground.

The 3D points to be associated with the trees are then removed from the 3D point cloud and an orthomosaic is formed. This then includes a view of the ground, including the shadows of the trees projected onto it.

It is also advantageous to identify the base points of the individual rooms that are contained in the 3D point cloud.

In order to be able to distinguish the cast shadows from other objects in the orthomosaics, the objects that can be recognized in the orthomosaics are classified. Using the orthomosaic, a classification is carried out in such a way that the cast shadows depicted therein are identified and the image points of the orthomosaic belonging to them are identified. To carry this out, an artificial neural network is preferably used. This is preferably trained using orthomosaics, in which a cast shadow classification for the image points is carried out manually. It has been shown that such a classification is possible with high reliability using a trained artificial neural network.

Based on the position of the sun, i.e. the azimuthal angle and elevation angle at which the sun shines on the earth at the time the aerial images were captured, it is possible to determine a shadow length that is associated with a height of 1.30 m above the base point. Using the azimuthal angle, a shadow direction can also be determined under which a shadow of a vertical object is oriented in a horizontal plane. The flatter the sun shines on the area, the longer the shadows cast by an object. The steeper the sun shines onto the terrain, the shorter the cast shadows are until the situation in which the sun is vertically above the trees at its zenith and no usable cast shadows are created. However, in medium geographical latitudes, such as those found in Central and Northern Europe as well as in North America, favorable conditions exist all day long in the winter months, under which the direction of solar radiation casts sufficient umbra shadows to be able to use the method and system described here.

Based on the shadow length determined from the position of the sun and the direction of irradiation, a width of the cast shadow can then be determined at a distance from the known and determined base of the tree or from an end of the cast shadow that faces the sun during detection. This is a measurement of the width of the tree trunk at chest height.

It becomes clear to the person skilled in the art that imaging conditions for the aerial images must be known in order to be able to relate the resolution of the image points to geometric dimensions. It is assumed that this information is determined during the evaluation and that the orthomosaic is created in such a way that the real distances corresponding to pixel distances in the orthomosaic are known or when evaluating and creating the orthomosaic from the structure-from-motion method can also be determined for the individual pixels. For this purpose, the aerial images are georeferenced using metadata. The position and time data are collected using a georeferencing device, preferably designed as a global satellite navigation system GNSS (Global Navigation Satellite System). This can also be referred to as a position and time determination unit.

With this system and the procedure implemented in it, it is possible to easily inventory large forest areas with regard to tree diameters at chest height. The cost of an unmanned aerial vehicle is affordable for most forest owners due to its rapidly increasing use. The evaluation can be implemented using software, in particular on a program-controlled computer, for example a desktop computer or a laptop. Due to the tremendous progress in computer development, the artificial neural network can also be used on a laptop or desktop computer be implemented. However, a centralized server is preferably used on which the trained neural network is implemented. The structure-from-motion evaluation and the other evaluation steps can also be carried out in a centralized computer.

Preferred Embodiments

• ein Luftfahrzeug mit einer Kamera zum Erfassen von Luftbildern eines Gebiets mit einem unbelaubten Baumbestand bei Sonnenschein, wobei das Luftfahrzeug eine Georeferenzierungseinheit zum Georeferenzieren der Luftbilder und zur Positionsbestimmung des Luftfahrzeugs umfasst;
• eine Auswerteeinrichtung, mit
  • einer 3D-Moddellierungseinheit, welche mittels eines Structure-from-Motion-Verfahrens eine 3D-Punktwolke des Gebiets ermittelt;
  • einer Reliefermittlungs- und Normalisierungseinheit, welche die 3D-Punkte hinsichtlich eines Oberflächenreliefs des Gebiets normalisiert;
  • eine Baumbestandteilentfernungseinheit, zum Erstellen eines hinsichtlich des Reliefs normalisierten Orthomosaiks der bodennahen 3D-Punkte des Gebiets;
  • einer Klassifizierungseinheit mit einem über Deep Learning trainierten künstlichen neuronalen Netz zum Ermitteln der Schlagschatten in dem normalisierten Orthomosaik;
  • einer Schattenlängenermittlungseinheit, die anhand des Sonnenstands bei der Erfassung die Schattenlänge für einen Baumdurchmesser auf Brusthöhe
  • ermittelt, und
  • einer Durchmesserermittlungseinheit, die für die ermittelten Schlagschatten deren Breite im Abstand der ermittelten Schattenlänge von einer dem Schlagschatten zuordenbaren Baumposition ermittelt und
  • einer Ausgabeinrichtung zum Ausgeben der Baumdruchmesser.

In particular, a system for determining tree diameters at breast height is created, which includes:

• an aircraft with a camera for capturing aerial images of an area with a bare tree population in sunshine, the aircraft comprising a georeferencing unit for georeferencing the aerial images and determining the position of the aircraft;
• an evaluation device, with
  • a 3D modeling unit, which uses a structure-from-motion method to determine a 3D point cloud of the area;
  • a relief determination and normalization unit that normalizes the 3D points with respect to a surface relief of the area;
  • a tree component removal unit for creating a relief-normalized orthomosaic of the ground-level 3D points of the area;
  • a classification unit with an artificial neural network trained via deep learning to determine the cast shadows in the normalized orthomosaic;
  • a shadow length determination unit, which determines the shadow length for a tree diameter at chest height based on the position of the sun
  • determined, and
  • a diameter determination unit which determines the width of the determined cast shadows at a distance of the determined shadow length from a tree position that can be assigned to the cast shadow and
  • an output device for outputting the tree diameters.

• Erfassen von Luftbildern eines Gebiets mit einem unbelaubten Baumbestand bei Sonnenschein mittels eines Luftfahrzeugs mit einer Kamera, wobei mittels einer Georeferenziereinheit zumindest die 3D-Position und die Zeit der Luftbilderfassung ermittelt und den Luftbildern als Metadaten zugefügt werden;
• Auswerten der Luftbilder umfassend die Teilschritte
  • Ermitteln einer 3D-Punktwolke des Gebiet mittels einer 3D-Moddellierungseinheit durch Anwenden eines Structure-from-Motion-Verfahrens;
  • Normalisieren der 3D-Punkte hinsichtlich eines Oberflächenreliefs des Waldgebiets mittels einer Reliefermittlungs- und Normalisierungseinheit;
  • Entfernen der Baumbestandteile und Erstellen eines hinsichtlich des Reliefs normalisierten Orthomosaiks der bodennahen 3D-Punkte des Gebiets mittels einer Baumbestandteilentfernungseinheit;
  • Klassifizieren der Bildpunkte des normalisierten Orthomosaiks zum Ermitteln der Schlagschatten mittels einer Klassifizierungseinheit mit einem über Deep Learning trainierten künstlichen neuronalen Netz;
  • Ermitteln einer Schattenlänge für einen Baumdurchmesser auf Brusthöhe mittels einer Schattenlängenermittlungseinheit, die anhand des Sonnenstands bei der Erfassung der Luftbilder die zu einem Lotrechten Gegenstand entsprechend der Brusthöhe auf einer flachen Ebene korrespondierende Schattenlänge ermittelt, und Ermitteln der Durchmesser mittels einer Durchmesserermittlungseinheit, indem für die ermittelten Schlagschatten deren Breite im Abstand der ermittelten Schattenlänge von der dem Schatten zuordenbaren Baumfußposition ermittelt wird; und
• Ausgeben der ermittelten Baumdurchmesser auf Brusthöhe.

Furthermore, a method for determining tree diameters at breast height is created, which includes the steps:

• Capturing aerial images of an area with a bare tree population in sunshine using an aircraft with a camera, at least the 3D position and the time of the aerial image capture being determined using a georeferencing unit and being added to the aerial images as metadata;
• Evaluating the aerial images including the sub-steps
  • Determining a 3D point cloud of the area using a 3D modeling unit by applying a structure-from-motion method;
  • Normalizing the 3D points with respect to a surface relief of the forest area using a relief determination and normalization unit;
  • Removing the tree components and creating a relief-normalized orthomosaic of the ground-level 3D points of the area using a tree component removal unit;
  • Classifying the image points of the normalized orthomosaic to determine the cast shadows using a classification unit with an artificial neural network trained via deep learning;
  • Determining a shadow length for a tree diameter at chest height by means of a shadow length determination unit, which uses the position of the sun when capturing the aerial images to determine the shadow length corresponding to a vertical object corresponding to the breast height on a flat plane, and determining the diameter by means of a diameter determination unit, in that for the determined cast shadows the width of which is determined at the distance of the determined shadow length from the tree foot position that can be assigned to the shadow; and
• Outputting the determined tree diameters at chest height.

The area is preferably a forest area. A forest area is defined as any area that has a population of trees.

The position of the end of the determined cast shadow facing the sun can be assumed as the assigned tree base position. Other embodiments determine the tree base position based on the 3D points of the 3D point cloud assigned to the tree. In such an embodiment, the relief determination and normalization unit can be designed to create and determine the root positions. A separate trunk position determination unit can also be provided for determining the trunk positions and thus the tree base positions.

In order not to have to collect the position of the sun when capturing the aerial images using the preferably unmanned aircraft, preferably only the geographical position and time are determined using the georeferencing unit, preferably a global satellite navigation system GNSS, and stored with them as metadata of the aerial images and can be used, for example. B. can be used in the evaluation using the structure-from-motion method. In a preferred embodiment, the position of the sun is then determined based on information about the geographical position and the time at which the aerial images were captured using astronomical data. The position of the sun can be calculated using a parameterized solar position curve or a parameterized path curve of the earth around the sun, taking into account the inclination of the earth's axis, or it can be retrieved from astronomical databases or it can be calculated based on the retrieved data.

In order to obtain precise information, in a preferred embodiment it is provided that the shadow length determination unit is designed to take into account the relief on which the shadow is cast when determining the shadow length. This information about the relief, which was originally determined using the structure-from-motion method, is used to adjust the lengthened or shortened shadows due to inclines or slopes in the terrain with regard to the normalization carried out in the orthomosaics. While a sloping terrain viewed from the side of the sun's rays causes the shadows to be shorter than on flat terrain, the shadows on a sloping terrain viewed from the irradiation side are lengthened compared to flat, horizontal terrain. These relief-related lengthenings and shortenings are preferably taken into account when determining the shadow length, which determines the distance at which the width of the tree shadow is measured from the base point.

In a particularly preferred embodiment, this determination is carried out individually for each tree. This means that the shadow determination unit is designed to determine the shadow length, which in the orthomosaic corresponds to a shadow of a vertical object with a height of 1.30 m, individually for each tree.

Since not only the shape of the terrain and the relief are important for the determination, but also vertical objects, namely the trees, are important, a preferred embodiment provides that the aerial images are not captured in nadir geometry, but rather preferably captured as oblique images. This means that the camera does not look straight down, but is pivoted laterally relative to the direction of flight and captures the aerial images preferably sideways, but alternatively also “looking forward” or “looking back”.

Due to the large number of parallax differences that arise when taking oblique images, vertical structures can be captured better and more precisely using oblique images within the structure-from-motion method.

The system therefore preferably has a trajectory determination unit which is designed to determine a trajectory for the aerial image capture, with which the aerial images of the area can be optimally captured. In a preferred embodiment, it is provided that the trajectory determination unit is designed to determine the trajectory for the aerial image capture by superimposing a regular linear or regular rectangular grid of reference points on the area and defining the trajectory so that it includes loops around the reference points, which are formed from semi-ellipses and semicircles, the aircraft being designed to fly along the flight path and to carry out the aerial image capture in such a way that each of the loops is assigned to one of the reference points and the aerial image capture is carried out in such a way that the reference point is always imaged in the center of the captured oblique images , which are recorded while flying the loop assigned to the reference point.

Particularly simple and favorable trajectories for data acquisition in terms of the time required and the resolution achieved, which optimally cover the area, provide that the semi-ellipses are constructed in such a way that their associated ellipses have a major semi-axis, the length of which corresponds to the distance between the loops corresponds to reference points flown around and have small semi-axes whose length corresponds to half the distance between the reference points and the reference points are each vertically below the contact point of the major and minor axis of the assigned semi-ellipse, and the semicircles have a radius that corresponds to the reference point distance, and centers of the semicircles lie vertically above the midpoints between adjacent reference points.

In other embodiments, it can be provided that the aerial images are captured in a vertical top view, ie in nadir geometry, and with a high overlap, for example 90% overlap. Here, a trajectory is preferably which includes parallel oriented straight sections.

Other embodiments may utilize a trajectory that includes parallel trajectory sections, with both oblique images and nadir images being captured during this. The additional oblique view images improve the capture of vertical structures.

Yet other embodiments can provide for an area to be flown over in a spiral pattern, with preferably a reference point, which is arranged in the middle of the spiral path or vertically below its center, being captured in the center of the images that are captured while the spiral trajectory is flying become.

The use of a convolutional neural network has proven to be particularly beneficial. H. a convolutional neural network has been proven to be an artificial neural network. This type of artificial neural network is particularly suitable for classifying local changes, such as those that occur at shadow edges, and the associated features.

To train the artificial neural network, tree component-free normalized orthomosaics with manually marked shadow areas are used.

It has proven to be particularly advantageous if the camera of the aircraft is designed to capture colored aerial images and the evaluation device is designed to carry out the evaluation for several color channels. This improves the robustness of the system and the determination method and thus in particular the precision of determining the tree diameters.

The features of the method have the same advantages as the corresponding features of the system.

• 1 eine schematische Ansicht zur Erläuterung einer Ermittlung eines Baumdurchmessers auf Brusthöhe;
• 2 eine schematische Darstellung eines Systems zur Ermittlung des Baumdurchmessers auf Brusthöhe;
• 3 ein schematisches Flussdiagramm eines Verfahrens zur Ermittlung des Baumdurchmessers auf Brusthöhe; und
• 4 eine schematische Darstellung einer Flugbahn für die Erfassung von Luftbildern.

The invention is explained in more detail below with reference to a drawing. Show here:

• 1 a schematic view to explain a determination of a tree diameter at breast height;
• 2 a schematic representation of a system for determining tree diameter at breast height;
• 3 a schematic flow diagram of a method for determining tree diameter at breast height; and
• 4 a schematic representation of a flight path for capturing aerial images.

In 1 A situation for determining tree diameters at chest height is shown schematically. A system 1000 includes, on the one hand, an aircraft 1100 with a camera 1150. The aircraft 1100 is preferably an unmanned aircraft, also known as an unmanned aerial vehicle (UAV).

The aircraft 1100 is used to fly over an area 300 with a tree population 350. The camera captures 1150 aerial images 400. Flying over area 300 takes place while tree stand 350 is bare of leaves. For example, flying over area 300 takes place at a time of year when the trees are not leafy.

Furthermore, the area is flown over while the sun 600 shines unobscured through clouds onto the trees 350.

The aircraft includes a georeferencing unit 1180, by means of which the 3D position of the aircraft 1100 is determined when the aerial images are captured. The time of recording is also determined. For georeferencing of the aerial images 400, the 3D position and the time when captured are stored as metadata with the aerial images.

The tree population 350 includes, for example, a tree 305. Due to flying in the sunshine, the sunlight 610 hits a tree trunk 310 of the tree 305 and is blocked by the tree trunk 310. Those areas of a floor 370 that are not illuminated by the sunlight 610 because this sunlight is blocked by the tree trunk 310 form an umbra 320, which is also referred to as a cast shadow. Image points 470 of the aerial image 400 representing the ground 370 thus have a clear contrast in the imaging area 420 of the umbra 320 compared to the remaining areas of the ground or the corresponding imaging areas 430 that are exposed to sunlight. The width of the cast shadow 325 corresponds to the tree diameter 315.

The area 300 is flown over by the aircraft 1100 in such a way that the aerial images 400 captured with the camera 1150 are suitable for creating a 3D point cloud of the objects captured in the aerial images and in particular of the ground 370 of the area using a structure-from-motion method to create.

The system 1000 for determining tree diameters at breast height is shown schematically in 2 shown. In 3 is a sequence of ver Steps for determining tree diameters at chest height are shown. These two figures provide the basis for the following description, which explains the determination of the tree diameter at breast height in more detail. The same technical features are marked with the same reference numbers in the figures. Reference numbers in the 1000 range refer to 2 . Reference numerals in the 2000 range refer to 3 , where equal hundreds of digits indicate that the corresponding device 1xyy of the system carries out the functions of the corresponding process steps 2xzz.

As already in connection with 1 As explained, the system 1000 includes an aircraft 1100 with a camera 1150 for capturing aerial images in direct sunlight of an area with a bare tree population 2100.

The position of the aircraft 1100 is preferably determined by means of satellite-based navigation, optionally supported by further information received via radio, which indicates a deviation from the position determined by satellite navigation, i.e. generally determined with a georeferencing unit 1110 (e.g. GNSS) and with the aerial images when captured saved 2110. The parameters of the camera, i.e. H. their orientation, imaging parameters etc. recorded and saved 2120.

The system 1000 also includes an evaluation device 1200 with which the captured aerial images are evaluated and the tree heights at chest height are determined.

Using the information contained in the aerial images and the information about the capture positions and capture times, it is possible to use a structure-from-motion method, as is known in the prior art, to determine a 3D point cloud that includes both points, which represent the tree components as well as the relief of the terrain of area 300.

This method is based on the fact that the same objects or points on the terrain are seen from different directions, i.e. H. different vertical angles and horizontal angles. Parallax differences arise between different aerial images, which can be used to draw conclusions about the position of the object/point of the terrain contained in both images in the real world. For this it is necessary to determine the position of the aircraft 1100 and the orientation of the camera 1150 at the times at which the images, i.e. H. the two aerial photos 400 are captured.

The evaluation device 1200 of the system 1000 therefore includes a 3D modeling unit 1300, which uses a structure-from-motion method to determine a 3D point cloud of the area 2300.

In order to simplify further evaluation, the 3D point cloud is normalized 2400. This is done in a relief determination and normalization unit 1400. This unit is designed to model a relief of the terrain of the area and then normalize the heights of the elements of the 3D point cloud. This is equivalent to “leveling” the relief of the terrain. 3D points or elements of the 3D point cloud, which represent, for example, the tips of various 20 m high tree trunks, all have a height value of 20 m after normalization, regardless of the relief of the terrain, i.e. H. regardless of whether the tree in question is in a hollow or on a slight hill.

The evaluation device 1200 further includes a tree component removal unit 1500. This is designed to remove from the 3D point cloud those elements or 3D points that belong to tree crowns and tree trunks 2500. Furthermore, in some embodiments, this unit is designed to determine the positions of base points of the Identifying tree trunks 2505.

The tree trunk removal unit 1500 creates from the remaining elements/points of the normalized 3D point cloud, i.e. H. of the ground-level 3D points or elements an orthomosaic of area 2510. In addition to the cast shadows, larger branches or similar lying on the ground may also be visible.

In order to identify the cast shadows, ie to classify the image points of the orthomosaic that can be assigned to a cast shadow of a tree, a classification unit 1600 is used. For this purpose, the evaluation device 1000 comprises a classification unit 1600, which classifies the image points or image point areas belonging to the shadows of trees in the orthomosaic 2600. The classification unit 1600 is by means of a trained artificial neural network 1610, preferably by means of an artificial neural convolutional network (English Convolutional Neural Network). realized. The artificial neural network can be trained, for example, with orthomosaics recorded and prepared in a similar way, in which the cast shadows are marked manually. This can be done, for example, using the software ArcGIS Pro Environmental Systems Research Institute, Inc. (ESRI), Redlands, California, USA.

The evaluation device 1200 also includes a cast shadow length determination unit 1700. This determines, based on the position of the sun at the time the aerial images are captured, on the basis of which the individual image points of the orthomosaic are determined, a shadow length that corresponds to a shadow cast by a 1.30 m high vertical object on a horizontal surface corresponds to 2700. Furthermore, the position of the sun can be used to determine in which direction the shadow is cast on a horizontal surface.

The evaluation device 1200 includes a diameter determination unit 1800 which, based on the information output by the cast shadow length determination unit 1700, determines the corresponding widths and thus trunk diameters of the trees causing the cast shadow in the classified cast shadow. The determination is carried out at a distance of a tree base that corresponds to the determined length of the cast shadow.

In some embodiments, the diameter determination unit 1800 measures the determined shadow length from one end of the classified cast shadow. In another embodiment, the tree component removal unit or another unit that evaluates the 3D point cloud determines the base positions of the trees. In a further development, the shadow lengths for cast shadows, which are correlated with a base point of a tree determined in this way, are plotted from the base point position determined.

Preferably, the diameter determination unit 1800 is also designed to output the corresponding tree diameters at chest height. In another embodiment, a separate output unit or output device 1900 can be provided in order to output the determined tree diameters at chest height 2900.

The tree diameter at chest height is output together with the coordinates of the associated tree. In addition, other data can also be output, such as the time at which the aerial photos were captured, information about uncertainties in the evaluation, etc.

In a further development of the invention, it is provided that the cast shadow length determination unit 1700 determines a length of the cast shadow for each tree or for each classified cast shadow, taking into account the relief at the corresponding location of the cast shadow 2710. This allows the information to be significantly improved. Depending on the slope of the terrain, the length of the cast shadow is increased or decreased. If the terrain on the side of a tree facing away from the sun rises, the length of the cast shadow is shortened. However, if the terrain on the side of the tree facing away from the sun slopes down, a cast shadow is extended. The relief shape therefore has an influence on the corresponding shadow length.

An improvement in the determination of vertical structures, in particular the 3D image points belonging to the tree trunk, results when the aerial images are captured as oblique images. In contrast to a classic acquisition of aerial images, in which the area is flown in parallel lines with a vertically oriented camera (nadir geometry), so that the aerial images have an coverage of around 90%, an improved flight pattern provides for the area to be given a grid is overlaid by reference points. This grid corresponds to crossing points of a preferably regular rectangular or square pattern formed from straight lines. The individual reference points, which are crossing points of the pattern formed from straight rectangular lines, are preferably flown around in a loop. In a particularly preferred embodiment, the system 1000 includes a trajectory determination unit 1105, which is designed to determine a trajectory 3000 for aerial image capture. The flight path 3000 is formed in a flight plane spaced from the ground of the area. The trajectory includes semicircular and semielliptical sections. 4 shows an example of such a trajectory 3000. Each reference point 3100 is assigned a semi-ellipse 3200. The reference point 3100 lies at the center of the ellipse, which defines the semi-ellipse 3200. A major semi-axis 3210 corresponds to the distance between adjacent reference points 3100. The semi-ellipses 3200 are connected to one another via semicircles 3300, the radii of which also correspond to the distance between adjacent reference points 3100. The center points 3310 of these semicircles 3300 are each located in the middle between two adjacent reference points 3100. This information relates to a core area of the area. Flying in can also take place via a semicircle 3300', the center of which 3310' is half a reference point distance “in front of” the first reference point 3100. While flying along the flight path 3000, the camera is now pivoted in such a way that the reference point 3100 flown around is imaged in the center of the captured aerial image. A change in camera orientation to the next reference point 3100 preferably takes place after flying through one half one of the semicircles 3300 of the flight path, ie at center positions 3320 of the semicircles 3300.

Preferably, the reference points 3100 are defined in a horizontal plane that coincides with the ground of the area in at least one point. This is preferably one of the reference points 3100. Although the flight plane in which the flight path is defined is spaced from the plane in which the reference points lie, these are in 4 superimposed on each other or projected into the drawing plane.

• Da die Schrägbilder, die bei diesem Verfahren erfasst werden, für eine Vielzahl von Parallaxen-Differenzen sorgen, ist es möglich nicht nur waagerechte Strukturen, sondern auch senkrechte Strukturen präzise zu modellieren. Daher können die Positionen und Bestandteile der Bäume zuverlässiger und vollständiger erfasst werden, was beispielsweise eine Ermittlung der Fußpunkte der Bäume und deren Positionen verbessert.

In addition to the semicircular and semielliptical sections, additional sections are added to the trajectory when “switching” from one row of the grid's reference points to the next row of the grid. The sections assigned to a reference point preferably result in a trajectory that is similar to a letter "ℓ" written in original cursive. The trajectory for a reference point arranged in a row resembles the typeface of handwritten "ℓ" letters arranged in a row:

• Since the oblique images captured with this method provide a large number of parallax differences, it is possible to precisely model not only horizontal structures, but also vertical structures. Therefore, the positions and components of the trees can be recorded more reliably and completely, which, for example, improves the determination of the base points of the trees and their positions.

Eliminating the elements of the 3D point cloud that are correlated with the trees is also improved.

Reference symbol list

300

Area

305

Tree

310

Tree trunk

315

Tree diameter

320

Drop shadow

325

Width of the drop shadow

350

trees

370

Floor

400

Aerial photos

420

Image area of the umbra

430

Image area of the illuminated floor

470

pixels

600

Sun

610

sunlight

1000

System for determining tree diameters at breast height

1100

aircraft

1105

Trajectory determination unit

1110

Georeferencing unit (e.g. GNSS)

1150

camera

1200

Evaluation device

1300

Modeling unit

1400

Relief determination and normalization unit

1500

Tree component removal unit

1600

Classification unit

1700

Cast shadow length determination unit

1800s

Diameter determination unit

1900

Output facility

2000

Method for determining tree diameters at breast height

2100

Capturing aerial images

2105

Determining the trajectory

2110

Saving positions and times of capture

2120

Saving the mapping parameters

2300

Detecting a 3D point cloud

2400

Normalize the 3D point cloud

2500

Removing tree parts

2505

Determining the tree bases

2510

Creating an orthomosaic of the points near the ground

2600

Classifying the image points with regard to the cast shadow affiliation

2700

Finding the shadow length of the chest height

2710

Taking the relief into account when determining the length of the shadow

2800

Determining shadow widths to determine tree diameters

2900

Output the tree diameter at chest height

3000

trajectory

3100

reference point

3200

Half ellipse

3210

major semi-axis

3220

minor semi-axis

3300, 3300`

Semicircle

3310, 3310`

Focus

3320

Center position

Claims (10)

Method (2000) for determining tree diameters at breast height comprising the steps: Capturing aerial images (400) of an area (300) with a bare tree population (350) in sunshine using an aircraft (1100) with a camera (1150), using a georeferencing unit (1110) to record at least the 3D position and the time of the aerial image capture determined and added to the aerial images (400) as metadata; Evaluating the aerial images (400) including the sub-steps Determining a 3D point cloud of the area (300) using a 3D modeling unit by applying a structure-from-motion method; Normalizing the 3D points with respect to a surface relief of the area (300) by means of a relief determination and normalization unit (1400); Removing the tree components and creating a relief-normalized orthomosaic of the ground-level 3D points of the area (300) using a tree component removal unit (1500); Classifying the image points of the normalized orthomosaic to determine the cast shadows using a classification unit (1600) with an artificial neural network trained via deep learning; Determining a shadow length for a tree diameter at chest height using a shadow length determination unit (1700), which uses the position of the sun when capturing the aerial images to determine the shadow length corresponding to a vertical object on a flat plane, and Determining the diameter by means of a diameter determination unit (1800) by determining the width of the determined cast shadows at a distance of the determined shadow length from the tree base position that can be assigned to the shadow; and Outputting the determined tree diameters at chest height. procedure (2000). Claim 1 , characterized in that the position of the sun is determined based on information about the geographical position and the time at which the aerial images were captured using astronomical data. procedure (2000). Claim 1 or 2 , characterized in that the shadow length determination unit is designed to take into account the relief on which the shadow is cast when determining the shadow length. procedure (2000). Claim 3 , characterized in that the shadow length is determined individually for each tree. Method (2000) according to one of the preceding claims, characterized in that the aerial images are captured as oblique images. procedure (2000). Claim 5 , characterized in that the trajectory for the aerial image capture is determined by means of a trajectory determination unit by superimposing a regular linear or regular rectangular grid of reference points (3100) on the area and the trajectory is determined so that it includes loops around the reference points (3100). , which are formed from semi-ellipses (3200) and semicircles (3300), and wherein when capturing the aerial images (400), the flight path (3000) is flown by means of the aircraft (1100) and the aerial image capture is carried out in such a way that each of the loops is one the reference points (3100) is assigned and the assigned reference point (3100) is always imaged in the center of the captured oblique images while the loop assigned to this reference point (3100) is flown. System (1000) for determining tree diameters at chest height, comprising: an aircraft with a camera (1150) for capturing aerial images of an area (300) with a bare tree population in sunshine, the aircraft (1100) having a georeferencing unit (1110) for georeferencing the Aerial images (400) and for determining the position of the aircraft (1100); an evaluation device, with a modeling unit (1300), which determines a 3D point cloud of the area (300) using the structure-from-motion method; a relief determination and normalization unit (1400), which normalizes the 3D points with respect to a surface relief of the forest area; a tree component removal unit (1500) for creating a relief-normalized orthomosakis of the ground-level 3D points of the area and determining the trunk positions; a classification unit (1600) with an artificial neural network trained via deep learning to determine the cast shadows in the normalized orthomosaic; a shadow length determination unit, which uses the position of the sun during detection and determines the shadow length for a tree diameter at chest height, and a diameter determination unit, which determines and outputs the width of the determined cast shadows at a distance of the determined shadow length from the tree base position that can be assigned to the shadow. System (1000) after Claim 7 , characterized in that the artificial neural network is a convolutional neural network. System (1000) according to one of the Claims 7 or 8th , characterized in that recorded tree component-free normalized orthomosaics with hand-marked shadow areas are used for training. System (1000) according to one of the Claims 7 until 9 , characterized in that the camera (1150) of the aircraft (1100) is designed to capture colored aerial images (400), and the evaluation device (1200) is designed to carry out the evaluation for several color channels.

Images