FIELD OF THE INVENTION

The invention relates to a device and method for determining a plant status. More in particular, the invention relates to a device and method for determining a height of an agricultural product.

BACKGROUND TO THE INVENTION

For many agricultural crops, it is desirable to determine the height of the plants, their biomass or other biophysical parameters noninvasively while the crop is growing. Possible applications are yield predictions before harvest as well as optimization of rate and timing of fertilizer and agrochemicals applications. For example, the biomass of an oilseed rate crop before winter is known to be a valuable indicator of nitrogen demand in springtime.

Measuring the height of other agricultural products, such as leafy vegetables, e.g. salad and kale, cereals, e.g. wheat, com, etc., is also important in order to follow the development of the crops and adjust farming operations such as fertilizer and crop protection products applications, irrigation.

In the case of grass for grazing, leaf production for grass is understood to be maximized by grazing to between 3.5 and 4 cm of residual height. By keeping the pasture in a growing state, a higher quality of grass will be produced in a green leafy base. Pre-grazing height can e.g. be between 8 and 9 cm, corresponding to three leaves approximately. If such grass is grazed down to between 3.5 and 4 cm, growth will be expected at 16,000 kg DM/ha.

Recently, electronic systems for determining crop height have been developed, which rely upon image acquisition for carrying out measurements. For example, WO2018/122237 discloses a device including a sensor (e.g. a mobile phone) arranged on a stick at a predetermined distance from the ground. A digital stereo image of part of a plant is recorded and the height of the plant may be determined from these images by means of image analysis. Although the prior art device often provides a satisfactory result, it would be desirable to have an even more versatile device. In particular, it would be desirable to avoid the stick and operate the device free-hand without the need for additional hardware. Other electronic systems for determining crop height are disclosed in documents W02013/087052 and DEI 02008009753. Document WO2013/0870520 discloses a device including a built-in stereo camera for determining a plant parameter. Such a stereo camera is however a dedicated device which is not present in most mobile devices like smartphones or tablets. Document DE 102008009753 discloses a dedicated ultrasound sensor for determining further plant parameters. It is therefore an object of the present invention to allow a user to carry out the determination with a state of the art camera, without stereoscopic capabilities, as present in mobile devices nowadays.

SUMMARY OF THE INVENTION

It is an objective of the present invention to provide an even more versatile method and device for measuring height of an agricultural product, such as oilseed rape. It will be appreciated that the method can also be used for determining a height of other agricultural products, such as kale, com (maize), grass or other crops.

According to a first aspect of the present invention, this and other objectives are achieved by a method for measuring a crop height of an agricultural product above ground using a portable device, positioning the portable device above the agricultural product, determining a first distance between the portable device and the ground based on a video feed acquired with a camera of the portable device, determining a second distance between the portable device and the agricultural product based on the video feed, and determining the crop height based on the determined first and second distances.

According to this approach, two distances are continuously determined from a video feed of the camera of the portable device. A first distance relates to the ground, while a second distance relates to the (top of) the agricultural product. The height of the product can then easily be determined as the difference between these distances (i.e. the first distance minus the second distance).

By the expression “video feed” is here intended a continuous stream of image frames acquired by the camera. The frame rate is typically at least five or at least ten frames per seconds (fps), and may, for a typical mobile phone, be e.g. 30 pfs or 60 fps. With this continuous measurement approach, a user may determine the height of the agricultural product on the move, by simply holding the device above the product. The use of a video feed allows the user to walk across a field in a continuous manner. According to the previously disclosed method (WO2018/122237), the user was required to stop while the device acquired images of the agricultural product. The present invention thus allows a faster measurement of a given plot of land or field.

In order to determine the distance to ground, it is necessary that the “bare” ground is at least occasionally visible between specimens of the agricultural product. The method is thus particularly useful for products which are not too dense.

According to one embodiment, each distance is acquired from a distance tracking module of the portable device, which distance tracking module is configured to continuously determine one or several distances between the portable device and objects in the field of view of the camera, based on the video feed. By identifying the ground and the agricultural product in the field of view (i.e. in the video image) the first and second distances may be identified.

Such a tracking module may estimate distances based on the contents of a plurality of consecutive images (frames) but may additionally include other sensors, such as an accelerometer, a gyroscope or a time-of-flight camera.

The tracking module may further be configured to store, at regular time intervals, determined distances in a designated memory area, from which the first and second distances may be acquired after identifying the ground and the agricultural product in the video image.

Such tracking modules and memory areas are known in the art, and one example, compatible with the present invention, is the Google ArCore® library. The Google ArCore library was originally developed for mobile phone applications using virtual reality. However, it was discovered that the same library could be used in the present invention.

In some embodiments, the method includes storing a record including data representative of the determined height in a memory of the portable device and/or in a database in communication with the portable device. Hence, the measured height is available for later inspection and/or use.

The method may also include determining a position of the portable device, using a built-in position determining unit, such as a GPS unit, and storing data representative of the position in the record. Hence, the record can contain data representative of the determined height of the agricultural product and the location at which the height was determined.

The height of the product may be determined at a plurality of locations, to create a pattern of product height across an area. In this case, the method may further include estimating a height of the agricultural product for additional positions within a geographical area based on the height determined at the plurality of measured positions. Such estimating can include using interpolation and/or extrapolation. The geographical area may be user selected. The height of the agricultural product can e.g. be interpolated between heights in measured locations. The height of the agricultural product between a measured position and the perimeter of the geographical area may e.g. be estimated by extrapolation. Alternatively, or additionally, the height of the agricultural product can be estimated by averaging of the individual determined heights.

According to a further aspect of the invention, a biophysical parameter of an agricultural product within an area may be estimated by determining the crop height of the agricultural product as described above, and calculating the biophysical parameter of the agricultural product based on the determined height. Example biophysical parameters include e. g. be biomass, leaf angle, leaf size, leaf area index, and canopy cover. The biophysical parameter, e.g. the dry matter biomass in kg or kg/ha, can e.g. be determined from the determined height by a factor determined by an empirical model. The calculation may be done in a processing unit of the portable device, or in a remote processing unit.

According to a further aspect of the present invention, this and other objectives are achieved by a portable device comprising a camera, and a processing unit configured to:

- determine a first distance between the portable device and ground above which the communication device is held based on a video feed acquired with the in camera,

- determine a second distance between the portable device and an agricultural product above which the communication device is held based on said video feed, and

- determine a height of the agricultural product based on the determined first and second distances. According to yet another aspect of the present invention, this and other objectives are achieved by a computer program product including software code portions configured to, when run on the processing unit of a portable device which has been positioned above ground above an agricultural product, execute a method according to the first aspect. The computer program product may be arranged to operate under e.g. the Android operating system, Apple IOS operating system, and/or Windows operating system.

It will be appreciated that any of the aspects, features and options described in view of the method apply equally to the portable device, system, support and computer program product and vice versa. It will also be clear that any one or more of the above aspects, features and options can be combined.

BRIEF DESCRIPTION OF THE DRAWING

The present invention will be described in more detail with reference to the appended drawings, showing currently preferred, example embodiments of the invention.

Figure 1 is a perspective view of a communication device held over rows of agricultural products.

Figure 2 is a schematic block diagram of a communication device according to an embodiment of the invention.

Figure 3 shows the device in figure 2 displaying a user interface according to an embodiment of the invention.

Figure 4 is a flow-chart of a method according to an embodiment of the invention.

Figure 5 shows the device in figure 1 displaying a current field of view of the camera.

The example embodiments are given by way of non-limitative illustration. It is noted that the figures are only schematic representations of embodiments of the invention that are given by way of non-limiting example.

DETAIFED DESCRIPTION

Figure 1 depicts an agricultural product in the form of plants 1 of a crop (e.g. com) planted in rows 2, with the underlying ground 3 visible between the rows 2. The height of the plants is indicted by h. A portable device 10 is shown in a position above the plants, at a first distance h _g to the ground, and at a second distance h _c to the crop, i.e. to the upper parts of the plants 1.

In figure 1, the portable device 10 is illustrated as “floating” in the air, but in reality it would of course be held by or carried by a user. Alternatively, the device 10 may be mounted to a support at a suitable distance from the ground. The support can e.g. be footwear, a walking stick, a user vehicle, a post, a gate, a fence, a mobile robot, a drone aerial vehicle, an animal, or the like. Optionally, the support may be configured to support the device at a plurality of different positions, i.e. at different distances from the ground. The different mounting positions can be associated with different measurement ranges. A first measurement range can e.g. for agricultural product heights of from 0 to 2 cm. A second measurement range can e.g. for agricultural product heights of from 1 to 30 cm. A third measurement range can e.g. for agricultural product heights of from 20 to 100 cm. However, other measurement ranges are possible.

The portable device 10 is here generally illustrated as a smartphone, but the components and properties of a portable device according to the invention will be discussed in greater detail with reference to figure 2. The device includes a camera 11 configured to acquire video feed (i.e. a sequence of image frames). The video feed may have a frame rate of 10 fps (frames per second) or more. As an example, the frame rate of a typical smartphone is 30 fps or 60 fps. The currently acquired video data may be referred to as the “field of view” of the camera. The device also comprises a display 12, e.g. a touchscreen, typically having an operating mode in which it is configured to display the field of view in real-time. The device also includes a processing circuitry (processor) 13 and a working memory 14.

The device is further provided with a distance tracking module 15 - possibly embodied by the processor 13 or as an additional (separate or embedded) circuitry - configured to determine a distance to a set of n points in the field of view. The set of n points (x, y and z coordinates) are continuously made available in a portion 14a of the device working memory 14. The identified points P may be indicated in the field of view (see figure 3). Such a distance tracking module 15 with associated memory 14a is known in the art, and one example is the Google ArCore ® library, running on the Android ® operating system. The distance estimation is performed primarily based on the video data, i.e. a continuous flow of image frames. In order to allow distance estimation based on such a flow of images, the images need to be slightly different, in other words the camera 11 needs to move at least slightly. Additional sensors 16, such as a gyro and/or accelerometer, may also be involved in the distance estimation process.

In addition, the device 10 here is provided with a GPS 19 or similar localization device, configured to provide the position of the device in world coordinates. The device 10 further preferably has a wireless communication interface 17, in communication with a remote database 18.

According to some embodiments, data representative of an area of land on which the height of the agricultural product is to be determined may be stored in a local storage (e.g. in the memory 14) or in the remote database 18. It is, for instance, possible to mark the area of land on a map on a screen, such as a touch-screen, of the portable device. Such marking can include drawing a perimeter contour around the area of land, e.g. on the map or aerial photograph. Alternatively, or additionally, it can be possible to demarcate the area of land by physically moving the portable device along the perimeter of the area of land, and storing locations, such as GPS coordinates, of the perimeter. Alternatively, or additionally, data representative of the perimeter of the area of land can be download by or uploaded to the portable device or the database. Hence, it is possible to identify an area of land, e.g. a predetermined pasture, in which the height of the agricultural product is to be determined. It will be appreciated that the above mentioned record(s) can include an indication of the area of land, e.g. a name and/or number.

According to an embodiment of the invention, the processing unit 13 is configured to perform a method according to the flow chart in figure 4. The method may be implemented by software code stored in the memory 14 and executed by the processing unit 13. The software may for example be embodied as an app running on the device operating system, e.g. Android ®, Apple IOS ® or Windows ®.

Figure 3 shows a device 10 displaying user interface of an embodiment of the invention. The user interface here includes a display area 23 and a lower portion 24 including one or more touch screen controls 25 allowing user interaction. As will be discussed below, one control may e.g. activate determining of a crop height h _c in a current location. Other controls may e.g. change the content of the display area or activate a calibration of the device. The display area 23 here displays a representation 20 of a geographical area, previously defined or selected and entered into local or remote storage as mentioned above. The graphical representation 20 may e.g. include landmarks such as roads 22, to help a user to identify the area in the real world. The representation 20 of the area of land may e.g. be shown in overlay on a map or aerial photograph.

With reference to figure 4, first, in step SI, the software is initiated, e.g. by clearing the working memory and launching (starting) the module 13 and memory 14, e.g. the ArCore ® library of an Android ® device. Next, in step S2, the device 10 is brought to a position above the rows 2 of plants, and kept in slight motion. The distance tracking module 15 will then automatically identify a set of n points (x, y) in the image plane for which a distance (z) from the image plane is determined. The distance z is determined from the video feed, i.e. the stream of image frames acquired by the camera. Additional input, from the sensors 16, may also be used. As the device 10 is moved, the points in the set are continuously tracked.

Figure 5 shows a fictive screen shot, showing an instant field of view with a plurality of points P _g , P _c marked therein. These points indicate coordinates (x, y) in the image plane for which the module 15 has estimated a distance (z). These points are also schematically indicated in figure 1. As is clear from figures 1 and 5, a first sub-set of points P _g are located on the ground between rows 2 of plants 1, while a second sub-set of points P _c are located on the plants 1 themselves.

In step S3, the program control awaits a user input to trigger the data acquisition. The user input can be a hardware button (e.g. volume control) or a touch screen based control 25 as discussed above. In response to receiving a trigger signal, program control continues to step S4, where a current camera position (X, Y, Z) is obtained from the localization system 19, and the current set of n points (x, y, z) are acquired (read out) from the ArCore ® library. As mentioned above, the coordinates x and y are taken to indicate the position in the image plane, and the coordinate z is taken to represent the normal distance from the image plane to the point.

In the next step S5, the set of n points is processed to determine a first distance d _g to the ground and a second distance d, to the crops (i.e. upper part of the plants 1).

According to one approach, the distance to the ground d _g and the distance to the crop d _c are calculated from percentiles of all points in the image, e. g: dg = P9o(Zi) dc = Pio(zi) where P _q (zd is the c/-th percentile of all z values in the image.

According to another possible approach, a subset of points most distant to the camera are chosen as points P _g located on the ground, while a subset of points most proximal to the camera are chosen as points P _c located on the upper parts of the plants. It may be advantageous to completely disregard extreme points, which may be considered to be caused by errors or interference.

As an alternative to the latter, or as an additional criterion, the color of the points may be used to distinguish ground points from crop points. For example, the (brown) ground points P _g can be separated from the (green) crop points P, through the RGB color values of the respective pixels.

The first and second distances d _g and d _c can then be determined as the average of the z-coordinates of the points P _g and P _c , respectively. If these z-coordinates are labelled z _gi and z _Ci , respectively, for i=l, ... ,N, where N is the number of points in each subset, d _g and d, are given by:

Optionally, each point also includes a confidence value, c. Using the same notations, d _g and d, are then given by:

In step S6, the crop height is determined by taking the difference between the distance to the ground and the distance to the crops, i.e. h _c = d _{g -} d, .

It is noted that it may not always be possible to determine a reliable crop height from the set of n points acquired form the ArCore library or similar module. For example, there may be an insufficient number of points P _g on the ground, or a too wide spread of distances to the crop. When a reliable measurement of the crop height has been performed, program control proceeds to step S7, to indicate the current location (X, Y, Z) on a representation 20 of a geographical area as a “measured location” L _m - A record containing the determined crop height h _c in association with the current location may also be stored in local memory 14 or in the remote database 18.

Steps S2 - S7 are repeated until it is determined in step S8 that the data collection procedure is finished, e.g. after receiving a further user input terminating the process. At this time, a plurality of measured locations L _m will be indicated in the display area 23, as shown in figure 5. It is noted that the various measured locations may be close together or far apart, depending on when the user triggers the data acquisition.

Continuing the process in figure 4, the illustrated method here includes a post processing step S9, occurring when the data collection is completed, e.g. when the user indicates that a sufficient number of locations have been measured. In this post processing step, average crop height and/or crop biomass is calculated for the geographical area indicated by representation 20.

If required or advantageous, an estimated crop height for additional locations LA, L _B may be determined using e.g. interpolation or extrapolation. For example, the crop height in a location LA between two measured locations L _m can be estimated by interpolation. Similarly, the crop height in a location L _B between a measured location and the boundary 21 of the geographical area 20 may be determined by extrapolation. Such additional locations may serve to improve the determination of average crop height.

Based on the determined average crop height of the geographical area, the processing unit 13 may be arranged to estimate a biophysical parameter of the agricultural product within the geographical area. For example, dry matter biomass, expressed e.g. in kg or kg /ha, can be determined by multiplying the determined height by an empirical factor. The empirical model used to determine the empirical factor may be specific for crop variety and growth stage.

In a final step S10, the results for the selected geographical area, e.g. including average crop height and/or estimated biomass can be displayed to the user in an appropriate manner. For example, the touch screen controls 25 may be used to change the display area to show relevant data results for the geographical area. When the height h _c of the agricultural product is determined on at least two moments in time, the growth and/or growth rate of the agricultural product can be determined. The growth and/or growth rate may indicate whether or not the agricultural product is sufficiently supplied with nutrients and water and may therefore be used to trigger fertilizer or water application.

Alternatively, or additionally, a nutritional status of the agricultural product can be determined from a leaf colour of the agricultural product. Since in the described method already at least one image of the agricultural product is taken, leaf colour can be determined from that image.

The portable device 10 may be in communication with an applicator system, such as a fertilizer system, irrigation system or the like. The nutritional status of the agricultural product or a need for a particular nutrient and/or water, or data representative thereof, can be communicated to the applicator system. Hence the water and/or nutrients can be supplied to the agricultural product on the basis of the estimated nutritional status. It will be appreciated that the nutritional status can be determined in dependence of a position within a geographical area 20, as described above for the height.

Herein, the invention is described with reference to specific examples of embodiments of the invention. It will, however, be evident that various modifications and changes may be made therein, without departing from the essence of the invention. For the purpose of clarity and a concise description features are described herein as part of the same or separate embodiments, however, alternative embodiments having combinations of all or some of the features described in these separate embodiments are also envisaged.

In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word ‘comprising’ does not exclude the presence of other features or steps than those listed in a claim. Furthermore, the words ‘a’ and ‘an’ shall not be construed as limited to ‘only one’, but instead are used to mean ‘at least one’, and do not exclude a plurality. The mere fact that certain measures are recited in mutually different claims does not indicate that a combination of these measures cannot be used to an advantage.