Method for determining soil texture

Abstract

The present invention relates to a method for determining soil texture, the method comprising: placing soil (32) and an aqueous liquid (34) in a substantially transparent container (16); mixing the soil and the aqueous liquid in the substantially transparent container; allowing the particles of the soil to separate into a plurality of layers (36a-c) due to gravity; capturing an image (38) of the layers in the substantially transparent container by moving a camera (12) of a computing device (10); and an app (30) of the mobile computing device detecting the layers in the captured image and determining the texture of the soil based on the detected layers.

Description

Technical Field

[0001] The present invention relates to a method for determining soil texture. The present invention also relates to a mobile computing device, a computer program product, a computer-readable storage medium, and an electrical signal. Background Art

[0002] Soil texture has a great impact on fertilizer and irrigation management.

[0003] For example, soil texture can be determined by laboratory analysis. In countries where soil chemistry / laboratory analysis is a routine test, soil texture is systematically determined. For this reason, extensive texture databases already exist in these countries. However, for small farmers, such information may be very scarce or non-existent.

[0004] Another way to determine soil texture involves shaking water and soil in a jar so that sand, silt, and clay separate due to gravity. This technique requires the user to measure the total height of the soil and the height of each layer (sand, silt, and clay) by eye to determine the percentage of each component. The user then compares the percentage of silt / sand / clay with the so-called soil texture triangle to determine the soil texture.

[0005] However, it may be difficult for the user to correctly measure the height, perform the percentage calculation, and / or interpret the soil texture triangle, which makes this technique (potentially) unreliable. Summary of the Invention

[0006] The object of the present invention is to overcome or at least mitigate the aforementioned problems.

[0007] According to a first aspect of the present invention, this object and other objects are achieved by a method for determining soil texture, the method comprising: placing soil and a water-based liquid in a substantially transparent container; mixing the soil and the water-based liquid in the substantially transparent container; allowing the particles of the soil to separate into multiple layers (segments) due to gravity; capturing an image of these layers in the substantially transparent container by a camera of a mobile computing device; and an app of the mobile computing device detecting these layers in the captured image and determining the texture of the soil based on the detected layers.

[0008] The present invention is based on the understanding that determining the soil texture at a specific location can be simplified and the accuracy of soil texture determination can be improved by using an app that captures an image of the layers using the camera of a mobile computing device and provides (using image analysis / recognition) the detection of the layers and determination of the soil texture on the mobile computing device. That is, farmers can simply use their smartphones (mobile computing devices) to capture an image, and then the app on the smartphone will automatically determine the soil texture. Farmers do not need to know how to calculate the aforementioned percentages and how to interpret the soil texture triangle, and potential errors in manually measuring the height of the layers, calculating percentages, and / or interpreting the soil texture triangle can be avoided. In addition, expensive equipment is not required because farmers may already own smartphones (i.e., mobile computing devices) on which the app can be downloaded.

[0009] The method for soil texture detection can use one or more of the edge and corner detection algorithms known per se in image processing (e.g., Sobel operator, Canny edge, Shi-Tomasi).

[0010] Detecting these layers in the captured image can include determining the thickness of each layer based on the captured image and determining the percentage of each layer based on the determined thickness. These layers typically include a (bottom) sand layer, a (middle) silt layer, and a (top) clay layer. In an example, the determined thickness of the sand layer is 410 pixels, the determined thickness of the silt layer is 390 pixels, and the determined thickness of the clay layer is 200 pixels, where the percentage of sand is 410 / (410 + 390 + 200) = 41%, the percentage of silt is 390 / (410 + 390 + 200) = 39%, and the percentage of clay is 200 / (410 + 390 + 200) = 20%, which corresponds to the soil texture "loam" (e.g., according to the soil texture triangle).

[0011] Detecting these layers in the captured image can include distinguishing at least one boundary (interface) between these layers by identifying the color difference between these layers. This can be achieved by looking at the different RGB values of the pixels in the image. Based on this, an algorithm can be trained to automatically distinguish (detect) the boundary or interface between two layers.

[0012] Detecting these layers in the captured image can include distinguishing at least one boundary (interface) between these layers by identifying the texture difference, especially the grain size difference, between these layers. This can include performing image processing to highlight the particles / roughness / grain size and using an algorithm that can extract the boundaries of each layer by being able to identify each layer. Generally, the sand layer has a coarser texture / grain size, the silt layer has a "medium" texture / grain size, and the clay layer has a finer texture / grain size.

[0013] Detecting layers in a captured image based on color and / or texture makes the layers easy to detect because color and texture are the main characteristics that define the layers.

[0014] The app can display a shape on the display of the mobile computing device that corresponds to at least a portion of the shape of the substantially transparent container, wherein the user positions and / or aims the mobile computing device such that the substantially transparent container seen in the live view on the display by the camera conforms to the displayed shape, and wherein the image of the layer in the substantially transparent container is captured when the mobile computing device is so positioned and / or aimed. In this way, an image of the layer can be captured at a predetermined angle relative to the substantially transparent container, which can in turn facilitate and / or improve the accuracy of detecting the layer in the captured image.

[0015] The substantially transparent container can include a funnel portion that converges into a cylindrical portion having a bottom. This can facilitate the placement of soil and a water-based liquid into the substantially transparent container via the funnel portion, while the percentage of the layer can be easily determined in the cylindrical portion. To this end, the volume of the soil can be equal to or less than the volume of the cylindrical portion of the substantially transparent container. The funnel portion can also provide an additional place to mix the soil and the water-based liquid, which can facilitate the mixing of, for example, heavy clay.

[0016] The method can further include the app providing a fertilizer recommendation based at least in part on the determined texture of the soil. The fertilizer recommendation can be displayed, for example, on the aforementioned display of the mobile computing device.

[0017] The mobile computing device can be a smartphone or a tablet computer. The smartphone or tablet computer can advantageously be a "general-purpose" smartphone or tablet computer, wherein the above functions are provided by an app stored on the smartphone or tablet computer.

[0018] The method can further include: using the emptied substantially transparent container as a rain gauge; capturing an image of the rainwater in the substantially transparent container by the camera of the mobile computing device; and the app determining the amount of rainwater in the substantially transparent container based on the captured image of the rainwater in the substantially transparent container. The advantage of doing so is that the substantially transparent container can have multiple uses, which is beneficial to the environment and avoids single-use plastics. In addition, using the app on the mobile computing device to determine the amount of rain provides several advantages compared to manual rain gauging. For example, push notifications can be conveniently sent to the mobile computing device via the app, prompting the user / farmer to check the substantially transparent container based on the weather forecast of rainfall. In another example, the determined amount of rainwater can be easily fed back from the app and the mobile computing device to the weather forecast provider.

[0019] The amount of rainwater in the substantially transparent container can be determined in the same way(s) as the layer discussed above, and the image of the rainwater in the substantially transparent container can also be captured at a predetermined angle relative to the substantially transparent container as also discussed above.

[0020] The substantially transparent container can be closed or closable by a lid having at least one of the color charts known to the app and a size known to the app, wherein the method can further include: capturing an image of an object with at least a portion of the lid as a background by the camera of the mobile computing device; and the app estimating the color of the object based on the captured image of the object considering the known color chart of the lid, and / or estimating the size of the object based on the captured image of the object considering the known size of the lid. In the former case, the color chart (e.g., red, green, and blue markings on the lid) should be visible in the captured image. In the latter case, when capturing the image, the object is preferably placed on the lid, and the known size (e.g., the diameter of the lid) should be visible in the captured image. When mixing soil and a water-based liquid in the substantially transparent container (which typically involves the user / farmer shaking the substantially transparent container and its contents), the lid is typically used to close the substantially transparent container, but the inventors have recognized that by programming the color chart and / or size of the lid into the app, the lid can also be conveniently used for color and / or size estimation. This (again) extends the use of the hardware, which is beneficial to the environment. Additionally, by using the app on the mobile computing device to estimate the color of an object, especially soil, the color estimation can advantageously be combined with the determined soil texture and GPS coordinates from the mobile computing device, and this data can, for example, be used to construct a soil color-texture map, i.e., a dataset that relates the texture-location-color of the soil. Here, if the farmer's location and the color of their soil (e.g., from a picture of their soil) are known, the soil texture at that particular farmer's location can be assumed.

[0021] The color chart known to the app can be red, green, and blue markings on the lid. Alternatively, for example, the color chart can be a color gradient or shades of different colors. For soil, the color chart can include different shades of black-brown-red-orange-yellow-white, but for practicality, there may be one shade of brown representing multiple brown shades - for example, ten brown shades (of the Munsell soil color chart), another shade of brown representing another ten brown shades, and so on.

[0022] Considering the known color chart of the lid, estimating the color of the object based on the captured image of the object may include: detecting the color difference between the color chart of the lid depicted in the captured image of the object and the known color chart of the lid, and using the detected difference as a calibration to estimate the (true) color of the object.

[0023] Once these percentages are determined, the app can determine the texture of the soil by querying the soil texture triangle that is programmed into the app. In other words, the app can determine the texture of the soil by applying the determined percentages of each layer to the soil texture triangle programmed into the app.

[0024] According to a second aspect of the present invention, there is provided a mobile computing device for determining soil texture, the mobile computing device including a camera, wherein the mobile computing device is configured to: instruct a user of the mobile computing device to capture an image of a soil layer in a substantially transparent container using the camera; detect the soil layers in the captured image of the soil layer in the substantially transparent container; and determine the soil texture based on the detected soil layers. This aspect may exhibit the same or similar features and technical effects as the first aspect, and vice versa. In particular, detecting the soil layers in the captured image may include: determining the thickness of each soil layer based on the captured image, and determining the percentage of each soil layer based on the determined thickness.

[0025] According to a third aspect of the present invention, there is provided a computer program product, the computer program product including computer program code that, when executed on a computer, performs the following steps: receiving an image of a soil layer in a substantially transparent container captured by a camera of a mobile computing device; detecting the soil layers in the received image; and determining the soil texture based on the detected soil layers. The computer program product may be a non-transitory computer program product. The computer may for example be the aforementioned mobile computing device. The computer program product is the aforementioned app. Additionally, the computer may be a remote computer (cloud solution) of the mobile computing device. This aspect may exhibit the same or similar features and technical effects as the first aspect and / or the second aspect, and vice versa. In particular, detecting the soil layers in the captured image may include: determining the thickness of each soil layer based on the captured image, and determining the percentage of each soil layer based on the determined thickness.

[0026] According to a fourth aspect of the present invention, there is provided a computer-readable storage medium including the computer program product according to the third aspect.

[0027] According to a fifth aspect of the present invention, there is provided an electrical signal implemented on a carrier and propagated on a dielectric, the electrical signal comprising a computer program product according to the third aspect.

[0028] According to a sixth aspect of the present invention, there is provided a mobile computing device, the mobile computing device comprising a computer-readable storage medium according to the fourth aspect.

[0029] According to a seventh aspect of the present invention, there is provided a mobile computing device according to the second aspect or the sixth aspect in combination with a substantially transparent container. Thus, the seventh aspect can be expressed as a system comprising a mobile computing device and a substantially transparent container. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] These and other aspects of the present invention will now be described in more detail with reference to the drawings which show currently preferred embodiments of the present invention.

[0031] Figure 1 Schematically shown is a mobile computing device and a container according to an aspect of the present invention.

[0032] Figure 2 is a flowchart of a method according to one or more embodiments of the present invention.

[0033] Figure 3 Shows an image of a soil layer in a substantially transparent container captured by a camera of a Figure 1 mobile computing device.

[0034] Figure 4 Shows a soil texture triangle.

[0035] Figure 5 Shows the use of a Figure 1 container as a rain gauge.

[0036] Figure 6 is a top view of the lid of the container. DETAILED DESCRIPTION

[0037] Figure 1 Stereoscopically shown is a mobile computing device 10 for determining soil texture according to an embodiment of the present invention.

[0038] The mobile computing device 10 can be handheld and / or portable. The mobile computing device 10 can be, for example, a smartphone (e.g., an iPhone or an Android phone) or a tablet computer (e.g., an iPad).

[0039] The mobile computing device 10 includes a (digital) camera 12. The mobile computing device 10 may also include one or more of an (electronic) display 14, a processor, a memory, a storage device, a GPS receiver, and a wireless communication device.

[0040] Figure 1 Also shown in a side view is a container 16 according to an embodiment of the present invention. The container 16 is a (substantially) transparent container 16. The (substantially) transparent container 16 is preferably close to glass in terms of transparency. The transparent container 16 can be made of a transparent plastic such as acrylic, for example. There is no need to provide a measurement scale on the container 16.

[0041] The transparent container 16 includes a funnel portion 18 and a cylindrical portion 20 having a bottom 22. The funnel portion 18 generally can have the shape of an (inverted) frustum of a cone. The funnel portion 18 has a free-end opening 24 that is wider (e.g., has a larger diameter) than the orifice 26 facing the cylindrical portion 20. The cylindrical portion 20 is preferably a straight cylindrical portion. The cylindrical portion 20 can have a volume in the range of 400 cm 3 to 600 cm 3 . In an example, the cylindrical portion 20 has a radius of 3 cm and a height of 18 cm, whereby the volume is approximately 509 cm 3 . The transparent container 16 is closed or closable by a lid (cover) 28 at the free-end opening 24.

[0042] Next, the mobile computing device 10 can be configured to perform various specific steps or actions (e.g., S5 - S7) detailed below via an app 30 (computer program product). The app 30 can be downloaded to the mobile computing device 10 via the aforementioned wireless communication means and stored on the aforementioned storage means. The app 30 can run or execute on the mobile computing device 10 using the aforementioned processor and memory.

[0043] Turning to Figure 2 , Figure 2 is a flowchart of a method according to one or more embodiments of the present invention.

[0044] At S1, the method includes placing soil 32 and a water-based liquid 34 in the transparent container 16. Specifically, a user (farmer or agronomist) can place a certain amount of soil 32 in the transparent container 16 and then add the water-based liquid 34 via the free-end opening 24. The volume of the soil 32 is preferably (slightly) less than the volume of the cylindrical portion 20 of the transparent container 16. In addition, the water-based liquid 34 can be, for example, water or water with salt (NaCl).

[0045] If the soil 32 is (too) dense, the mobile computing device 10 / app 30 can instruct the user to dry and then grind the soil sample 32 before placing it in the transparent container 16.

[0046] At S2, the method includes mixing soil 32 and a water-based liquid 34 in a transparent container 16. This can specifically include closing the transparent container 16 with a lid 28, and then the user / farmer shakes the transparent container 16 and its contents 32, 34 until all soil particles are suspended in the water-based liquid 34.

[0047] Then, at S3, the transparent container 16 can be left standing on a flat surface for a period of time, such as at least two hours or about one day, allowing the particles of soil 32 to separate into multiple layers (segments) 36a-c due to gravity. These layers typically include a bottom sand layer 36a, a middle silt layer 36b, and a top clay layer 36c. The (remaining) water-based liquid 34 is located above the clay layer 36c.

[0048] At S4, the method includes capturing an image 38 of the layers 36a-c in the transparent container 16 by moving the camera 12 of the computing device 10. Prior to this, the mobile computing device 10 / app 30 can indicate to the user to capture such an image 38, for example, by displaying an appropriate message on the display 14. Here, the app 30 can also display a shape 40 corresponding to at least a part of the shape of the transparent container 16 on the display 14, where the user positions and / or aims the mobile computing device 10 such that the transparent container 16 seen in the live view on the display 14 by the camera 12 conforms as well as possible to the displayed shape 40 (as Figure 1 shown), and where the image 38 is captured when the mobile computing device 10 is so positioned and / or aimed. The shape 40 can be, for example, the profile of the transparent container 16 when viewed from the side, i.e., the line marking the outer boundaries of the transparent container 16, or the reverse silhouette of the transparent container 16. The shape 40 can be applied as an (overlay) filter in the live view on the display 14.

[0049] Then, the app 30 detects the layers 36a-c in the captured image 38 at S5 and determines the texture of the soil 32 based on the detected layers 36a-c at S6. The determined texture of the soil 32 (soil texture classification) can be displayed on the display 14, such as "loamy sand", "loam", "silt", "sandy clay loam", etc.

[0050] With the present invention, the determination of soil texture at a specific location can be simplified relative to the fully manual techniques described in the background section of the present application, and the accuracy of soil texture determination can also be improved. That is, the user / farmer can simply use their smartphone 10 to capture an image 38, and then the app 30 will automatically determine the soil texture. The farmer does not need to know how to calculate the percentages of layers 36a-c and how to interpret the soil texture triangle, and potential errors in manually measuring the heights of layers 36a-c, calculating the percentages, and / or interpreting the soil texture triangle can be avoided. In addition, no expensive equipment is required because the farmer may already own a smartphone 10 on which the app 30 can be downloaded.

[0051] For further reference Figure 3 , detecting the layers 36a-c in the captured image 38 may include: determining the thickness 42a-c of each layer 36a-c based on the captured image 38, and determining the percentage of each layer 36a-c relative to the total amount of the soil 32 based on the determined thickness 42a-c. In an example, the determined thickness 42a of the sand layer is 410 pixels, the determined thickness 42b of the silt layer is 390 pixels, and the determined thickness 42c of the clay layer is 200 pixels. Among them, the percentage of sand is 410 / (410 + 390 + 200) = 41%, the percentage of silt is 390 / (410 + 390 + 200) = 39%, and the percentage of clay is 200 / (410 + 390 + 200) = 20%. Once the percentages are determined, the app 30 can determine the texture of the soil 32, for example, by querying the soil texture triangle, which can be programmed into the app 30. In this example, the percentages of 41% sand, 39% silt, and 20% clay correspond to the soil texture "loam". An example of the soil texture triangle is as Figure 4 shown.

[0052] Specifically, detecting the layers 36a-c in the captured image 38 may include: differentiating the dividing lines 44a-b between the layers 36a-c by identifying the color and / or texture (grain size) differences between the layers 36a-c. In addition, the upper boundary 44c of the top layer 36c can be detected by identifying the color and / or texture differences between the layer 36c and the (residual) water-based liquid 34. In addition, the lower boundary 44d of the bottom layer 36a can be detected by directly detecting the bottom 22 of the transparent container 16 and considering the material thickness of the bottom 22.

[0053] At S7, the method can further include the app 30 providing a fertilizer recommendation at least partially based on the determined texture of the soil 32. For example, on sandy soil, the fertilizer recommendation may be to apply fertilizer in small amounts more frequently to avoid leaching. In clay, the fertilizer recommendation can be to apply fertilizer less frequently but in larger amounts. The fertilizer recommendation can be displayed, for example, on the display 14 of the mobile computing device 10.

[0054] Go to Figure 5 , the transparent container 16, which is emptied and has no lid 28, can further be used as a rain gauge. In this use, the transparent container 16 can be placed, for example, in the field. Here, an image 46 of the rainwater 48 in the transparent container 16 can be captured by the camera 12 of the mobile computing device 10, where the app 30 determines the amount of rainwater 48 in the transparent container 16 based on the captured image 46. The amount of rainwater 48 in the transparent container 16 can be determined in the same (s) way as the above-described layers 36a-c. The amount of rainwater 46 can be, for example, the volume of the rainwater or the height H expressed in mm. The volume of the rainwater can be calculated based on the height and volume or (base) area of the cylindrical portion 20, which can be known to the app 30.

[0055] In addition, as shown at 50, a push notification can be conveniently sent to the mobile computing device 10 via the app 30 (and using the wireless communication means), thereby prompting the user / farmer to check the transparent container 16 based on the past rainfall weather forecast. In addition, as shown at 52, the determined amount of rainwater can be easily automatically fed back from the app 30 and the mobile computing device 10 to the weather forecast provider via the wireless communication means.

[0056] To provide an accurate fertilizer recommendation, for example, at S7, it may be an advantage to know the exact amount of rainwater that has fallen on the field. Fields with sufficient rainwater may require more fertilizer to achieve high yields. Excessive rainwater can also cause nutrient loss, that is, too much water in the field may absorb the nutrients present in the soil and carry them out of the field. Similarly, the app 30 can take this into account when communicating the fertilizer recommendation to the user.

[0057] Next to Figure 6, the lid 28 of the transparent container 16 can have a color chart known to the app 30. The color chart can be, for example, a red marker 54a, a green marker 54b, and a blue marker 54c on the lid 28. Here, an image 56 of the object 58 can be captured by moving the camera 12 of the computing device 10 with at least a portion of the lid 28 as the background and with the markers 54a-c visible, wherein the app 30 can consider those known colors of the lid 28 and estimate the color of the object 58 based on the captured image 56. Estimating the color of the object 58 can include: detecting the color difference between the color chart depicted in the captured image 56 and the known color chart of the lid 28, and using the detected difference as a calibration to estimate the (true) color of the object 58. For example, if the imaged marker 54c becomes bluer than specified, the app 30 can reduce the blue component of the imaged color of the object 58 to determine or estimate the true color of the object 58. The object 58 can be, for example, soil (such as soil 32), whereby the color estimate can be combined with the determined soil texture and the corresponding GPS coordinates (given by the GPS receiver of the mobile computing device 10). This data can in turn be used to construct a soil color-texture map.

[0058] Alternatively or complementarily, the lid 28 for the transparent container 16 can have a size known to the app 30. The known size can be, for example, the diameter D of the lid 28. Here, an image (such as image 56) of an object (such as object 58) placed on the lid 28 can be captured by moving the camera 12 of the computing device 10 with at least a portion of the lid 28 as the background and with the known size D visible, wherein the app 30 can consider the known size D of the lid 28 and estimate the size of the object 56 based on the captured image. For example, given a known size D = 10 cm, if the imaged object 58 has a width of 600 pixels and the imaged diameter D is 800 pixels, the estimated size (width) of the object 56 is (600 / 800) * 10 cm = 7.5 cm.

[0059] Those skilled in the art will recognize that the present invention is in no way limited to the above-described embodiments. On the contrary, many modifications and variations are possible within the scope of the appended claims.

[0060] In addition, using the container as a rain gauge (such as in claim 10) and using the lid to estimate the color and / or size of an object (such as in claims 11, 12, 13) can both be implemented independently of the determination of soil texture.

Claims

1. A method for determining soil texture, comprising: placing soil (32) and a water-based liquid (34) in a transparent container (16); mixing the soil and the water-based liquid in the transparent container; allowing the particles of the soil to separate into multiple layers (36a-c) due to gravity; capturing an image (38) of these layers in the transparent container by moving a camera (12) of a computing device (10); and an app (30) of the mobile computing device detecting these layers in the captured image and determining the texture of the soil based on the detected layers, wherein detecting these layers in the captured image includes determining the thickness (42a-c) of each layer based on the captured image and determining the percentage of each layer based on the determined thickness.

2. The method according to claim 1, wherein, detecting these layers in the captured image includes distinguishing at least one dividing line (44a-b) between these layers by identifying the color difference between these layers.

3. The method according to any one of the preceding claims, wherein, detecting these layers in the captured image includes distinguishing at least one dividing line (44a-b) between these layers by identifying the texture difference between these layers.

4. The method according to any one of claims 1-2, wherein, detecting these layers in the captured image includes distinguishing at least one dividing line (44a-b) between these layers by identifying the grain size difference between these layers.

5. The method according to claim 1, wherein, these layers include a sandy soil layer (36a), a silt soil layer (36b), and a clay soil layer (36c).

6. The method according to claim 1, wherein, the app displays a shape (40) corresponding to at least a part of the shape of the transparent container on a display (14) of the mobile computing device, wherein the user positions and / or aims the mobile computing device such that the transparent container seen in the live view on the display by the camera conforms to the displayed shape, and wherein the image of these layers in the transparent container is captured when the mobile computing device is so positioned and / or aimed.

7. The method according to claim 1, wherein, the transparent container includes a funnel portion (18) that converges into a cylindrical portion (20) having a bottom (22).

8. The method according to claim 1, further comprising: the app providing a fertilizer recommendation based at least in part on the determined texture of the soil.

9. The method according to claim 1, wherein, the mobile computing device is a smart phone or a tablet computer.

10. The method according to claim 1, further comprising: using the emptied transparent container as a rain gauge; capturing an image (46) of rainwater (48) in the transparent container by the camera of the mobile computing device; and the app determining the amount of rainwater in the transparent container based on the captured image of the rainwater in the transparent container.

11. The method according to claim 1, wherein, The transparent container is closed or closable by a lid (28) having at least one of the color charts (54a-c) known to the app and a dimension (D) known to the app.

12. The method according to claim 11, further comprising: capturing an image (56) of an object (58) with at least a portion of the lid as a background by a camera of the mobile computing device; and the app estimating the color of the object based on the captured image of the object considering the known color chart of the lid, and / or estimating the dimension of the object based on the captured image of the object considering the known dimension of the lid.

13. The method according to claim 12, wherein estimating the color of the object based on the captured image of the object considering the known color chart of the lid includes: detecting a color difference between the color chart of the lid depicted in the captured image of the object and the known color chart of the lid, and using the detected difference as a calibration to estimate the color of the object.

14. The method according to claim 1, wherein once these percentages are determined, the app determines the texture of the soil by querying a soil texture triangle programmed into the app.

15. A mobile computing device (10) for determining soil texture, the mobile computing device including a camera (12), wherein the mobile computing device is configured to: instruct a user of the mobile computing device to capture an image (38) of soil layers (36a-c) in a transparent container (16) using the camera; detect the soil layers in the captured image of the transparent container; and determine the soil texture based on the detected soil layers, wherein detecting the soil layers in the captured image includes: determining the thickness (42a-c) of each soil layer based on the captured image, and determining the percentage of each soil layer based on the determined thickness.

16. A computer program product (30) including computer program code which, when executed on a computer (10), performs the following steps: receiving an image (38) of soil layers (36a-c) in a transparent container captured by a camera (12) of a mobile computing device (10); detecting the soil layers in the received image; and determining the soil texture based on the detected soil layers, wherein detecting the soil layers in the captured image includes: determining the thickness (42a-c) of each soil layer based on the captured image, and determining the percentage of each soil layer based on the determined thickness.

Images