Raw wood analyzing system and raw wood analyzing method

Abstract

To accurately identify a core material and a sapwood of a raw wood without depending on the skill of a worker.SOLUTION: A raw wood analyzing system 10 includes a photographing means 14 for photographing a end 84 of a raw wood 80, an infrared measuring means 22 for measuring a reflectance of infrared rays from the end 84, and an analyzing means 30 for generating a reflectance map of infrared rays of the end 84 from an image photographed by the photographing means 14 and a measurement result of the infrared measuring means 22, and determining a boundary position between heartwood and sapwood in the end 84 based on the reflectance map. As a result, since it is not necessary for the operator to distinguish between the heartwood and the sapwood, the heartwood and the sapwood can be accurately identified by the analysis means 30 regardless of the skill level of the operator even for a tree type that is difficult to visually identify.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a log analysis system and a log analysis method for analyzing logs before they are sawn. [Background technology]

[0002] Logs cut in the forest and transported to a sawmill are piled up in a log yard. They must then be graded according to their length, diameter, heartwood, and sapwood size to maximize the yield of lumber produced after sawing. While the grading method for logs varies depending on the cut, as shown in Figure 5(a), in order to produce straight-grained lumber 96, which offers particularly high added value, the sapwood 88 and heartwood 92 (colored portion) are identified from the cut surface (hereinafter referred to as the "end grain") 84 of the log 80. Furthermore, to determine the size of the straight-grained lumber 96 that can be harvested, the dimensions of the sapwood 88 are measured, as shown in Figure 5(b), for example.

[0003] Many tree species have a dark core (heartwood 92) surrounded by a light-colored ring (sapwood 88). The sapwood 88 is often white, often referred to as sapwood, while the heartwood 92 is often reddish-brown, often referred to as reddish-meat. In Japanese cedar, one of Japan's most representative conifers, the heartwood 92 can be reddish-brown (called redheart) or black (called blackheart). This is because the sapwood 88 contains living tissue, whose growth rings allow water to pass from the roots to the leaves. Meanwhile, the heartwood 92 lacks living tissue and contains substances such as polyphenols, gums, and resins that darken the wood, preventing water from passing through from the roots. Therefore, in tree species with color differences between the sapwood and heartwood, such as those described above, the sapwood 88 and heartwood 92 can be distinguished visually. Recently, there is a technology that photographs the end grain 84 with a camera and uses the color differences in the image to distinguish between sapwood 88 and heartwood 92, identify the outline of the heartwood 92 (see, for example, Patent Document 1), determine the center of the tree rings (see, for example, Patent Document 2), and detect the position of the pith (see, for example, Patent Document 3).

[0004] On the other hand, there are some species in which the color of the heartwood 92 is not very different from the color of the sapwood 88. Examples include radiata pine, cypress, fir, and hemlock. For tree species in which there is little color difference between the heartwood 92 and sapwood 88, it is difficult to distinguish them based on color differences using camera image processing. In such cases, skilled operators grade the logs 80 in the log yard to extract high-value-added straight-grained lumber 96 from the sapwood 88. Specifically, skilled operators line up the logs 80 delivered from the forest to the sawmill's log yard and measure the length of each one. Next, the operator visually inspects the end grain 84 of the log 80 to distinguish between the sapwood 88 and the heartwood 92, and measures the width of the sapwood 88 in three or four locations (see, for example, Figure 5(b)). Based on these lengths and the width of the sapwood 88, a grading chart for the log 80 (see, for example, Figure 6) is consulted, and a spray is applied to the cutting position, and the grade is spray-marked on the end grain 84. The chainsaw operator then cuts the logs at the spray position along the length, which are then stacked into log grades according to length and size of the sapwood 88, and then each item is sent to the sawmill. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Laid-Open No. 2009-248320 [Patent Document 2] Japanese Patent Application Laid-Open No. 2011-088436 [Patent Document 3] Japanese Patent Application Laid-Open No. 2017-040548 Summary of the Invention [Problem to be solved by the invention]

[0006] In the above process, there are skilled operators who distinguish and measure the sapwood 88 from the heartwood 92 in the log yard, and operators who cut the logs to length with a chainsaw at the spray-on position. In such cases, the accuracy of grading, particularly the accuracy of distinguishing between the sapwood 88 and the heartwood 92, varies depending on the operator's level of skill, which can result in fluctuations in the yield of straight-grained lumber 96. Furthermore, the work involves an environment in which safety must be taken into consideration, such as when working with a chainsaw, when logs 80 are piled in the log yard, and when heavy machinery is running. The present invention has been made in view of the above-mentioned problems, and an object of the present invention is to accurately distinguish between the heartwood and sapwood of a log without depending on the skill level of the operator. [Means for solving the problem]

[0007] (Aspects of the invention) The following embodiments of the present invention are examples of the configuration of the present invention, and are described in terms of sections to facilitate understanding of the various configurations of the present invention. Each section does not limit the technical scope of the present invention, and while taking into consideration the detailed description of the invention, the technical scope of the present invention may also include the replacement or deletion of some of the components of each section, or the addition of other components.

[0008] (1) A log analysis system including: a photographing means for photographing the butt end of a log; an infrared measuring means for measuring the infrared reflectance from the butt end; and an analysis means for generating an infrared reflectance map of the butt end from the image photographed by the photographing means and the measurement results of the infrared measuring means, and for determining the boundary position between the heartwood and sapwood at the butt end based on the reflectance map.

[0009] The log analysis system described in this section analyzes logs before they are sawn, and includes a photographing means, an infrared measuring means, and an analyzing means. The photographing means photographs the buttocks (cut surfaces) of the logs, and the infrared measuring means measures the infrared reflectance from the buttocks of the logs photographed by the photographing means. The analyzing means then generates an infrared reflectance map of the buttocks using the image of the buttocks photographed by the photographing means and the measurement results of the infrared measuring means. That is, the analyzing means determines the shape of the buttocks from the image of the buttocks, extracts the measurement result portion for the buttocks based on the shape of the buttocks from the measurement results of the infrared measuring means, and generates a map of the infrared reflectance distribution for the buttocks.

[0010] In standing trees, moisture is absorbed from the roots and transported to the leaves through the sapwood, but not through the heartwood. Therefore, although the moisture content of logs varies by tree species, it tends to be low in the heartwood and high in the sapwood. Furthermore, because infrared light is absorbed by water at specific wavelengths, the reflectance of infrared light obtained by irradiating the buttocks is thought to correlate with the moisture (moisture content) inside the buttocks and logs. Therefore, by generating an infrared reflectance map for the buttocks as described above, it is expected that the difference in moisture content between the heartwood and sapwood will appear on the map. Therefore, the analysis means determines the boundary between the heartwood and sapwood at the buttocks based on the generated reflectance map. This eliminates the need for the operator to distinguish between the heartwood and sapwood of the log. Therefore, even for tree species that are difficult to distinguish visually, the analysis means can accurately distinguish between the heartwood and sapwood, regardless of the operator's level of skill. In other words, regardless of the species of the log or the color difference between the sapwood and heartwood, the system accurately distinguishes between sapwood and heartwood by utilizing the relative difference in infrared reflectance between the two. Therefore, logs can be graded accurately, ensuring a yield equivalent to that of experienced workers. Furthermore, the system reduces the time and effort required for workers to grade logs in the log yard, improving safety and production efficiency.

[0011] (2) In the above paragraph (1), the photographing means includes a depth camera that photographs the buttocks and measures the depth of objects in the photographed image, and the analysis means determines the contour of the buttocks in the photographed image from the photographed image and depth information obtained from the depth camera, and uses the determined contour of the buttocks to specify the range for generating the reflectance map. This is a log analysis system. In the log analysis system described in this section, the imaging means includes a depth camera, which captures an image of the end grain of the log and measures the depth of objects in the captured image. Therefore, the depth camera can obtain depth information (distance from the depth camera) of the end grain and surrounding objects within the imaging range of the depth camera.

[0012] The analysis means acquires the captured image and associated depth information from the depth camera and determines the contour of the buttock in the captured image. That is, based on the acquired captured image and depth information, the analysis means extracts a portion whose depth information matches the estimated distance from the depth camera to the buttock and whose shape matches the estimated contour shape of the buttock, and determines this as the contour of the buttock. The analysis means then uses the determined contour of the buttock to identify the range for generating a reflectance map for the buttock. This accurately identifies the range to be generated as a reflectance map of the buttock, thereby improving the efficiency of reflectance map generation and further improving the accuracy of distinguishing heartwood from sapwood.

[0013] (3) In the log analysis system according to the above item (2), the analysis means uses relative infrared reflectance within the specified range when generating the reflectance map. In the log analysis system described in this section, when the analysis means generates an infrared reflectance map of the buttocks, it uses the relative infrared reflectance within the range for generating the infrared reflectance map, specified as described in section (2) above. That is, although there is a difference in moisture content between the heartwood and sapwood of a log, the moisture content of each is not always constant. Therefore, by using the relative infrared reflectance within the range considered to be the buttocks for which the infrared reflectance map should be generated, which is specified using the contours of the buttocks, the system generates an infrared reflectance map without relying on the magnitude of the infrared reflectance itself, which correlates with the moisture content of the heartwood and sapwood. This makes the difference between the infrared reflectance of the heartwood and the infrared reflectance of the sapwood more pronounced, allowing for more accurate identification of heartwood and sapwood.

[0014] (4) In the above item (2), the analysis means determines the vertical and horizontal end points of the end point in the photographed image from the determined contour of the end point, and calculates the vertical and horizontal widths of the end point based on the depth information of each determined end point. In the log analysis system described in this section, the analysis means calculates the vertical and horizontal widths of the buttock. Specifically, the analysis means determines the two vertical and two horizontal end points of the buttock in the captured image from the contour of the buttock determined as described in section (2) above. The analysis means then calculates the distance between the two vertical end points of the buttock based on depth information associated with those two end points, and uses this as the vertical width of the buttock. Similarly, the analysis means calculates the distance between the two horizontal end points of the buttock based on depth information associated with those two end points, and uses this as the horizontal width of the buttock. This allows the vertical and horizontal widths of the buttock, in other words, the diameter of the log, to be accurately determined without the need for an operator to perform measurements, improving the accuracy of log grading.

[0015] (5) In the log analysis system according to the above item (4), the analysis means calculates the size of the heartwood and the sapwood based on the determined boundary position, the determined contour of the buttock, and the depth information. In the log analysis system described in this section, the analysis means calculates the size of the heartwood and sapwood at the butt end. That is, once the boundary position between the heartwood and sapwood determined as described in section (1) above and the outline of the butt end determined as described in section (2) above are determined, points for calculating size, such as the upper and lower ends of the heartwood or the upper and lower ends of the sapwood portion located above the heartwood, are extracted. Therefore, the analysis means calculates the distance between the points for calculating size based on the depth information associated with those points. This allows the size of the heartwood and sapwood to be accurately determined without the need for an operator to perform measurements, further improving the accuracy of log grading.

[0016] (6) A log analysis system according to the above item (5), wherein the analysis means outputs instructions for cutting positions of the log based on the content of the analysis. In the log analysis system described in this section, the analysis means further outputs instructions for cutting positions of the log based on the results of analysis by the analysis means, such as the boundary position between the heartwood and sapwood, the outline of the butt end, the vertical and horizontal widths of the butt end, and the sizes of the heartwood and sapwood. That is, the analysis means outputs instructions for appropriate cutting positions that will maximize the yield of lumber obtained from the log, based on the grade of the log determined as a result of the analysis. As a result, the instructions output by the analysis means are referenced and the logs are cut at appropriate positions by, for example, an operator of the cutting line equipment or a cutting saw, thereby consistently ensuring a high yield.

[0017] (7) A log analysis system according to the above paragraph (1), further comprising a storage means for storing the analysis results of the analysis means, including the reflectance map, wherein the analysis means includes an artificial intelligence unit for analyzing the reflectance map and determining the boundary position, and the artificial intelligence unit performs additional learning using the reflectance map stored in the storage means. The log analysis system described in this section further includes a storage means for storing the analysis results obtained by the analysis means, including an infrared reflectance map of the buttocks. The analysis means also includes an artificial intelligence unit, which is responsible for determining the boundary position between the heartwood and sapwood. That is, the artificial intelligence unit is trained to analyze the infrared reflectance map of the buttocks generated by the analysis means and determine the boundary position between the heartwood and sapwood. Furthermore, the artificial intelligence unit is configured to perform additional learning to determine the boundary position between the heartwood and sapwood, using the infrared reflectance map of the buttocks stored in the storage means. This allows the artificial intelligence unit to more accurately distinguish between heartwood and sapwood, and even if the boundary position between the heartwood and sapwood has an irregular shape, the artificial intelligence unit that performs additional learning can easily determine the boundary position.

[0018] (8) A log analysis method that photographs the end grain of a log, measures the infrared reflectance from the end grain, generates an infrared reflectance map of the end grain from the photographed image of the end grain and the measurement results of the infrared reflectance, and determines the boundary position between the heartwood and sapwood at the end grain based on the reflectance map.

[0019] (9) A log analysis method in which, when photographing the end grain, the depth of objects in the photographed image is further measured, the contour of the end grain in the photographed image is determined from the photographed image and the measured depth information, and the determined contour of the end grain is used to identify the range for generating the reflectance map. The log analysis methods described in (8) and (9) are executed using the log analysis systems described in (1) and (2) above, respectively, and thus have the same effects as those of the log analysis systems described in (1) and (2) above. [Effects of the Invention]

[0020] Because the present invention has the above-described configuration, it is possible to accurately distinguish between the heartwood and sapwood of a log, regardless of the skill level of the worker. [Brief explanation of the drawings]

[0021] [Figure 1] 1 is a block diagram showing an example of the configuration of a log analysis system according to an embodiment of the present invention, together with a log to be analyzed. [Figure 2] FIG. 1 is a flow chart showing an example of a procedure of a log analysis method according to an embodiment of the present invention. [Figure 3] FIG. 10 is an image diagram showing a color map as an example of a reflectance map. [Figure 4] 1 is a graph showing an example of moisture content distribution of a log. [Figure 5] (a) is an image of the butt end showing an example of the structure of a log, and (b) is an image showing a simplified view of the measurement points of the sapwood. [Figure 6] 1 is an example of a grading table used when grading logs. DETAILED DESCRIPTION OF THE INVENTION

[0022] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. Here, detailed descriptions of parts that are the same as or corresponding to those in the prior art will be omitted, and the same reference numerals will be used throughout the drawings to indicate the same or corresponding parts. Figure 1 shows a schematic diagram of an example of the configuration of a log analysis system 10 according to an embodiment of the present invention, with a log 80 to be analyzed shown on the left side of Figure 1. Log analysis system 10 is assumed to be installed in, for example, a sawmill, and the log 80 to be analyzed has been cut down in a forest and transported to the sawmill, and may be either undebarked or debarked.

[0023] As shown in FIG. 1 , a log analysis system 10 according to an embodiment of the present invention includes a photographing unit 14, an infrared measuring unit 22, an analyzing unit 30, and a storage unit 50. The photographing unit 14 photographs the butt end 84 of a log 80, and in this embodiment, includes a depth camera 18. The depth camera 18 photographs the butt end 84 and measures the depth of objects in the photographed image of the butt end 84. For example, a device combining a LiDAR and a visible light camera is used. Depth here refers to the distance between the depth camera 18 and the photographed object, and such depth information is acquired for each pixel in the photographed image of the butt end 84. To ensure accurate depth measurement, the depth camera 18 is preferably installed so that the distance and angle of the depth camera 18 relative to the butt end 84 meet the specifications required by the device constituting the depth camera 18, and the center of the butt end 84 is preferably positioned directly in front of the depth camera 18. The photographed image of the end grain 84 obtained by the depth camera 18 and the depth information associated with the photographed image are transmitted to the analysis means 30.

[0024] The infrared measuring means 22 measures the infrared reflectance from the buttock 84 of the log 80 and, in this embodiment, is comprised of an infrared camera 26. The infrared camera 26 receives infrared light reflected from objects within its imaging range and measures the infrared reflectance within the imaging range. The infrared camera 26 may, for example, use infrared light in the wavelength range of 8 to 14 μm. However, this is merely an example. As described below, infrared light in any wavelength range other than 8 to 14 μm may be used if the characteristic of water, which absorbs infrared light at specific wavelengths, can be utilized. Furthermore, to reduce the influence of sunlight and other factors, a camera with a narrower wavelength range or a pass filter that passes only specific wavelengths may be used. Furthermore, imaging using the infrared camera 26 may be performed in an environment less susceptible to the influence of sunlight and other factors. The position of the infrared camera 26 relative to the buttock 84 is determined to meet the specifications required by the equipment constituting the infrared camera 26. The infrared reflectance obtained by the infrared camera 26 is transmitted to the analysis means 30. The infrared camera 26 may also measure the intensity of the reflected infrared light and transmit it to the analysis means 30, which may then calculate the reflectance of the infrared light based on the measured intensity.

[0025] The analysis means 30 executes various analyses performed in the log analysis system 10 and, in this embodiment, also controls the entire log analysis system 10. The analysis means 30 of this embodiment includes a contour determination unit 32, a reflectance map generation unit 34, an artificial intelligence unit 38, a dimension calculation unit 42, an instruction output unit 46, and a data storage unit 48. The contour determination unit 32 determines the contour of the buttock 84 in the captured image based on the image acquired from the depth camera 18 of the imaging means 14 and the associated depth information; this will be described in detail later. The reflectance map generation unit 34 generates an infrared reflectance map 60 (see FIG. 3 ) of the buttock 84 based on the image captured by the depth camera 18 of the imaging means 14 and the measurement results of the infrared camera 26 of the infrared measurement means 22. In this embodiment, the contour of the buttock 84 determined by the contour determination unit 32 and the measurement results of the infrared camera 26 are used; the generation method will be described in detail later.

[0026] The artificial intelligence unit 38 analyzes the infrared reflectance map 60 of the butt grain 84 generated by the reflectance map generation unit 34 to determine the boundary position between the heartwood 92 and the sapwood 88 (see FIG. 5 ) at the butt grain 84. For this reason, the artificial intelligence unit 38 is trained to determine such boundary positions from the reflectance map 60. For example, training data is generated by manually labeling the heartwood 92 and the sapwood 88 with bounding boxes for the infrared reflectance map 60 of the butt grain 84 prepared in advance. Then, such training data is trained based on, for example, a YOLO (You Only Look Once) algorithm, and the resulting artificial intelligence is used as the artificial intelligence unit 38 to distinguish between the heartwood 92 and the sapwood 88. The artificial intelligence unit 38 is also configured to perform additional training to determine the boundary position between the heartwood 92 and the sapwood 88 using the reflectance map 60 stored in the storage means 50, as described below.

[0027] The dimension calculation unit 42 calculates the vertical and horizontal widths (diameters) of the butt end 84 and the sizes of the heartwood 92 and sapwood 88 based on the contour of the butt end 84 determined by the contour determination unit 32, the boundary position between the heartwood 92 and sapwood 88 determined by the artificial intelligence unit 38, and the depth information acquired by the depth camera 18. The calculation method will be described in detail later. The instruction output unit 46 outputs instructions to, for example, a worker at a sawmill based on the results of the analysis by the analysis means 30. In this embodiment, it outputs instructions regarding the cutting position of the log 80. The analysis content used by the instruction output unit 46 may include, depending on the instructions to be output, the contour of the butt end 84 determined by the contour determination unit 32, the boundary position between the heartwood 92 and sapwood 88 determined by the artificial intelligence unit 38, the sizes of each part of the log 80 calculated by the dimension calculation unit 42, and the depth information acquired by the depth camera 18.

[0028] The data storage unit 48 outputs various analysis results, including the infrared reflectance map 60, analyzed by the analysis unit 30, the captured images and depth information acquired by the depth camera 18 of the imaging unit 14, and the infrared reflectance acquired by the infrared camera 26 of the infrared measurement unit 22, to the storage unit 50 for storage. The storage unit 50 is configured to store the above data output by the data storage unit 48 and is constructed, for example, as a database. This database may be, for example, a document-oriented database, and can accommodate future additions of items. The data in the database may be viewed or downloaded from a management screen accessed via a web server or the like.

[0029] The log analysis system 10 according to an embodiment of the present invention is not limited to the configuration shown in FIG. 1 . For example, some components may be deleted or modified, or other components may be added, depending on the circumstances. Furthermore, the components of the analysis means 30 are divided into functional units, not into the software or hardware that actually constitutes the analysis means 30. Any software or hardware may be used for the software and hardware that constitute the analysis means 30. Furthermore, the storage means 50 may be incorporated into the hardware or software that constitutes the analysis means 30. While components requiring data communication are connected by lines in FIG. 1 , the actual connection method between them may be arbitrary, including wired or wireless, and an appropriate communication standard may be used depending on the data. The imaging means 14 and the infrared measurement means 22 may also have direct access to the storage means 50, so that their acquired data is stored in the storage means 50 without going through the data storage unit 48.

[0030] Next, a log analysis method according to an embodiment of the present invention, which utilizes the log analysis system 10 shown in Figure 1, will be described along the flow of the flow diagram shown in Figure 2. Please refer to Figures 1 and 5 as appropriate for the configuration of the log analysis system 10 and the log 80 to be analyzed. Note that the flow diagram shown in Figure 2 shows only an example of the procedural flow, and the log analysis method according to the embodiment of the present invention is not limited to the flow diagram of Figure 2. For example, depending on the configuration and situation of the log analysis system 10, some of the steps shown in Figure 2 may be deleted, changed, rearranged, or added as appropriate.

[0031] S10 (measurement of log length): Measure the length of the log 80 to be analyzed. The length may be measured by an operator in the same manner as in the past, or may be measured automatically on the line. S20 (End grain photography): The depth camera 18 of the photography unit 14 and the infrared camera 26 of the infrared measurement unit 22 photograph the end grain 84 of the log 80 to be analyzed. At this time, the depth camera 18 and the infrared camera 26 are positioned so that the end grain 84 of the log 80 is in a position suitable for photography. For example, the log 80 may be temporarily stopped midway along a conveyance line in a sawmill and photography may be performed in that state. Alternatively, the log 80 to be analyzed may be elevated using a cylinder or the like to prevent objects other than the log 80 from being positioned at a distance similar to the distance from the depth camera 18 to the end grain 84. Furthermore, to reduce the effects of direct sunlight, wind, rain, and the like, the depth camera 18 and the infrared camera 26 may be placed in some kind of enclosure to stabilize the photography environment. After photography is completed, the obtained data (photographed image, depth information, infrared reflectance) is transmitted to the analysis unit 30 and stored in the storage unit 50 via the data storage unit 48.

[0032] S30 (End grain contour determination): The contour determination unit 32 determines the contour of the end grain 84 in the captured image based on the captured image obtained from the depth camera 18 in S20 and the depth information associated with it. For example, using a set distance from the depth camera 18 to the end grain 84 at the time of capture, parts of the captured image whose depth information does not match the set distance are excluded from the analysis. Furthermore, using the expected shape and size of the end grain 84, parts that do not match these are excluded from the analysis. In this way, parts that match the set distance and the shape and size of the end grain 84 are extracted and used as the contour of the end grain 84. Note that as long as the contour of the end grain 84 can be determined in the above manner, the specific algorithm for contour determination may be any configuration.

[0033] S40 (Reflectance Map Generation): The reflectance map generator 34 generates an infrared reflectance map 60, such as that shown in FIG. 3, using the infrared reflectance obtained from the infrared camera 26 in S20. First, only the portion corresponding to the end grain 84 is cut out from the image captured by the infrared camera 26 using the outline of the end grain 84 determined in S30. In other words, the outline of the end grain 84 is used to identify the range for generating the infrared reflectance map 60. The infrared reflectance sensing results within the identified range are then converted into image data for visualization, so that the reflectance sensing results are reflected relatively within the range. Furthermore, the image may be converted into a color map image, for example, by coloring areas with low infrared reflectance blue and areas with high infrared reflectance red. In this manner, the infrared reflectance map 60 of the end grain 84 is generated.

[0034] S50 (boundary position determination): The artificial intelligence unit 38 analyzes the infrared reflectance map 60 of the butt grain 84 generated in S40 to determine the boundary position between the heartwood 92 and the sapwood 88 at the butt grain 84. FIG. 4 shows the moisture content distribution at the butt grain 84 of Radiata pine, an example of a tree species in which the color of the heartwood 92 and the color of the sapwood 88 do not appear to differ at the butt grain 84. For example, moisture absorbed from the roots of a standing tree passes through the sapwood 88 and not the heartwood 92. This tendency is clearly shown in the graph in FIG. 4. In other words, in FIG. 4, the moisture content is high in the portion considered to be the sapwood 88, which is close to the periphery of the log 80, and low in the portion considered to be the heartwood 92, which is far from the periphery of the log 80. Considering these characteristics and the property of water to absorb infrared rays at specific wavelengths, it is thought that the infrared reflectance at the end grain 84 reflects the difference in moisture content between the sapwood 88 and the heartwood 92.

[0035] Therefore, the artificial intelligence unit 38 is trained to analyze the infrared reflectance map 60, distinguish between areas with high and low infrared reflectance, and determine the boundary position between the heartwood 92 and the sapwood 88 from the reflectance map 60. For example, in the infrared reflectance map 60 of FIG. 3, a dark-colored area 62 with high infrared reflectance appears in the center as a roughly circular shape, surrounded by light-colored areas 64 with low infrared reflectance. Therefore, the artificial intelligence unit 38 relatively distinguishes the dark-colored area 62 with high infrared reflectance as corresponding to the heartwood 92, and the light-colored area 64 with low infrared reflectance as corresponding to the sapwood 88. The artificial intelligence unit 38 then determines the area between the dark-colored area 62 and the light-colored area 64 in the reflectance map 60 as the boundary position between the heartwood 92 and the sapwood 88.

[0036] S60 (dimension calculation): The dimension calculation unit 42 calculates the dimensions of necessary locations that can be determined from the buttock 84 of the log 80. For example, the dimension calculation unit 42 calculates the diameter of the buttock 84 of the log 80 and the width of the sapwood 88 at the buttock 84. When calculating the diameter of the log 80, the dimension calculation unit 42 determines the vertical and horizontal endpoints of the buttock 84 in the captured image from the outline of the buttock 84 in the image captured by the depth camera 18, as determined in S30 above. This determines the top and bottom endpoints and the left and right endpoints of the buttock 84, and then calculates the distance between the top and bottom endpoints of the buttock 84 and the distance between the left and right endpoints of the buttock 84. Specifically, based on the depth information associated with each endpoint, the x, y, and z coordinates of each endpoint in three-dimensional space are obtained, and the distance is calculated using the three-dimensional Euclidean distance formula. That is, the distance between the upper and lower end points of butt end 84 is calculated as described above and is used as the vertical width (vertical diameter) of butt end 84, and the distance between the left and right end points of butt end 84 is calculated as described above and is used as the horizontal width (horizontal diameter) of butt end 84.

[0037] Furthermore, when calculating the width of the sapwood 88, the dimension calculation unit 42 uses the outline of the buttock 84 determined in S30 and the boundary position between the heartwood 92 and the sapwood 88 determined in S50. For example, when calculating the width A of the sapwood 88 shown in FIG. 5(b), the dimension calculation unit 42 determines the upper end point of the buttock 84 from the outline of the buttock 84, and further determines the upper end point of the heartwood 92 from the boundary position between the heartwood 92 and the sapwood 88. Then, as in the calculation of the diameter, the dimension calculation unit 42 obtains the x, y, and z coordinates of the upper end points of the buttock 84 and the heartwood 92 in three-dimensional space based on the depth information associated with these two end points, and calculates the distance between the upper end point of the buttock 84 and the upper end point of the heartwood 92 as the width A of the sapwood 88 using the three-dimensional Euclidean distance formula. Widths B, C, and D of the sapwood 88 in FIG. 5(b) are also calculated in the same manner using the end points of the buttock 84 and the heartwood 92. The size (diameter) of the heartwood 92 can be easily calculated from the diameter of the butt end 84 and the width of the sapwood 88.

[0038] S70 (Log Grade Determination): The instruction output unit 46 grades the log 80 to be analyzed based on the length of the log 80 determined in S10 above, the width of the sapwood 88 calculated in S60 above, etc. For example, grading is performed based on whether the measured length of the log 80 and the calculated widths of three points of the sapwood 88 satisfy the criteria of a grading table such as that shown in FIG. S80 (Instruction output): Based on the analysis results of the target log 80 analyzed up to this point, the instruction output unit 46 outputs instructions for the cutting position of the log 80 to the operator or cutter of the cutting line equipment. The operator or cutter of the cutting line equipment cuts the analyzed log 80 according to the output cutting position instructions. Note that a mechanism that automatically cuts the log 80 based on the cutting position instructions output in this step may be incorporated into the log analysis system 10.

[0039] S90 (save analysis results): The data storage unit 48 stores the analysis results of the target log 80 that have been analyzed up to this point in the storage means 50. At this time, all of the analysis results may be stored, or only some of the analysis results may be stored as needed. S100 (artificial intelligence additional learning): The artificial intelligence unit 38 performs additional learning to determine the boundary position between the heartwood 92 and sapwood 88, using the infrared reflectance map 60 of the end grain 84 stored in the storage means 50. That is, the heartwood 92 and sapwood 88 portions of the infrared reflectance map 60 stored in the storage means 50 are labeled with bounding boxes by a human or other person, and the artificial intelligence unit 38 is then made to perform additional learning using these labels. The log analysis method according to this embodiment of the present invention is completed with the steps up to this point.

[0040] According to the embodiment of the present invention configured as described above, the following effects can be achieved. That is, as shown in Fig. 1, a log analysis system 10 according to the embodiment of the present invention analyzes a log 80 before sawing, and includes a photographing means 14, an infrared measuring means 22, and an analyzing means 30. The photographing means 14 photographs an end grain 84 of the log 80, and the infrared measuring means 22 measures the infrared reflectance from the end grain 84 of the log 80 photographed by the photographing means 14 (see S20 in Fig. 2). The analyzing means 30 then uses the image of the end grain 84 photographed by the photographing means 14 and the measurement results of the infrared measuring means 22 to generate an infrared reflectance map 60 of the end grain 84 (see Fig. 3) (see S40 in Fig. 2). That is, the analysis means 30 grasps the shape of the butt grain 84 from the image of the butt grain 84, extracts the measurement result portion for the butt grain 84 from the measurement results of the infrared measurement means 22 based on the shape of the butt grain 84, and generates a map 60 of the infrared reflectance distribution for the butt grain 84.

[0041] 4 and 5, in a standing tree, moisture is absorbed from the roots and transported to the leaves through the sapwood 88, but not through the heartwood 92. Therefore, although the moisture content of the log 80 varies depending on the tree species, the moisture content tends to be low in the heartwood 92 and high in the sapwood 88. In addition, because infrared rays are absorbed by water at specific wavelengths, the reflectance of infrared light obtained by irradiating the buttock 84 is thought to correlate with the moisture (moisture content) in the buttock 84 and inside the log. Therefore, by generating the infrared reflectance map 60 for the buttock 84 as described above, it is expected that the difference in moisture content between the heartwood 92 and the sapwood 88 will appear on the map 60. Therefore, the analysis means 30 determines the boundary position between the heartwood 92 and the sapwood 88 at the buttock 84 based on the generated reflectance map 60 (see S50 in FIG. 2). This eliminates the need for the worker to distinguish between the heartwood 92 and sapwood 88 of the log 80. This allows the analysis means 30 to accurately distinguish between the heartwood 92 and sapwood 88, regardless of the worker's level of skill, even for tree species that are difficult to distinguish visually. That is, regardless of the tree species of the log 80 or the color difference between the sapwood 88 and the heartwood 92, the sapwood 88 and the heartwood 92 can be accurately distinguished by utilizing the relative difference in infrared reflectance between the sapwood 88 and the heartwood 92. Therefore, the log 80 can be graded accurately, ensuring a yield equivalent to that achieved by a skilled worker. Furthermore, the labor required for workers to grade in the log yard is reduced, improving safety and production efficiency.

[0042] In addition, in the log analysis system 10 according to the embodiment of the present invention, as shown in FIG. 1, the imaging means 14 includes a depth camera 18. This depth camera 18 captures an image of the butt 84 of the log 80 and measures the depth of objects in the captured image. Therefore, the depth camera 18 can acquire depth information (distance from the depth camera 18) of the butt 84 and its surrounding objects within the imaging range of the depth camera 18. The analysis means 30 acquires the captured image and the depth information associated therewith from the depth camera 18, and determines the contour of the butt 84 in the captured image from them (see S30 in FIG. 2). That is, based on the acquired captured image and depth information, the analysis means 30 extracts a portion whose depth information matches the estimated distance from the depth camera 18 to the butt 84 and whose shape matches the estimated contour shape of the butt 84, and determines that portion as the contour of the butt 84. Then, the analysis means 30 uses the contour of the end grain 84 determined in this way to specify the range for generating the reflectance map 60 for the end grain 84. This makes it possible to accurately specify the range for generating the reflectance map 60 for the end grain 84, thereby improving the efficiency of generating the reflectance map 60 and further improving the accuracy of distinguishing between the heartwood 92 and the sapwood 88.

[0043] Furthermore, in the log analysis system 10 according to the embodiment of the present invention, when the analysis means 30 generates the infrared reflectance map 60 of the buttock 84, it uses the relative infrared reflectance within the range for generating the infrared reflectance map 60, as specified above. That is, although there is a difference in moisture content between the heartwood 92 and the sapwood 88 of the log 80, the moisture content is not always constant. Therefore, by using the relative infrared reflectance within the range that is considered to be the buttock 84 for which the infrared reflectance map 60 is to be generated, which is specified using the contour of the buttock 84, the infrared reflectance map 60 is generated without relying on the magnitude of the infrared reflectance itself, which correlates with the moisture content of the heartwood 92 and the sapwood 88. This makes the difference between the infrared reflectance of the heartwood 92 and the infrared reflectance of the sapwood 88 more pronounced, making it possible to more accurately distinguish between the heartwood 92 and the sapwood 88.

[0044] Furthermore, in the log analysis system 10 according to the embodiment of the present invention, the analysis means 30 calculates the vertical and horizontal widths of the buttock 84 (see S60 in FIG. 2). Specifically, the analysis means 30 determines the two vertical and two horizontal end points of the buttock 84 in the captured image from the contour of the buttock 84 determined as described above. The analysis means 30 then calculates the distance between the two vertical end points of the buttock 84 based on the depth information associated with these two end points, and uses this as the vertical width of the buttock 84. Similarly, the analysis means 30 calculates the distance between the two horizontal end points of the buttock 84 based on the depth information associated with these two horizontal end points, and uses this as the horizontal width of the buttock 84. This allows the vertical and horizontal widths of the buttock 84, in other words the diameter of the log 80, to be accurately determined without the operator having to perform measurements, thereby improving the accuracy of grading the log 80.

[0045] Additionally, in the log analysis system 10 according to the embodiment of the present invention, the analysis means 30 calculates the sizes of the heartwood 92 and sapwood 88 at the butt end 84 (see S60 in FIG. 2). That is, once the boundary position between the heartwood 92 and sapwood 88 and the outline of the butt end 84 are known, it is possible to extract points from which to calculate the size, such as the upper and lower ends of the heartwood 92 or the upper and lower ends of the portion of the sapwood 88 located above the heartwood 92. Therefore, the analysis means 30 can calculate the distance between the points from which to calculate the size based on the depth information associated with those points. This allows the sizes of the heartwood 92 and sapwood 88 to be accurately determined without the operator having to perform measurements, thereby further improving the accuracy of grading the log 80.

[0046] Furthermore, in the log analysis system 10 according to the embodiment of the present invention, the analysis means 30 outputs instructions for cutting positions of the log 80 based on the results of the analysis by the analysis means 30, such as the boundary position between the heartwood 92 and sapwood 88, the outline of the butt grain 84, the vertical and horizontal widths of the butt grain 84, and the sizes of the heartwood 92 and sapwood 88 (see S80 in FIG. 2). That is, the analysis means 30 outputs instructions for appropriate cutting positions that will maximize the yield of lumber products obtained from the log 80, based on the grade of the log 80 determined as a result of the analysis. This allows, for example, an operator of the cutting line equipment or a cutting saw to cut the log 80 at an appropriate position by referring to the instructions output from the analysis means 30, thereby ensuring a stable high yield.

[0047] Furthermore, as shown in Fig. 1, the log analysis system 10 according to the embodiment of the present invention further includes a storage means 50 for storing the analysis results by the analysis means 30, including an infrared reflectance map 60 of the buttocks 84. The analysis means 30 also includes an artificial intelligence unit 38, which is responsible for determining the boundary position between the heartwood 92 and the sapwood 88. That is, the artificial intelligence unit 38 is trained to analyze the infrared reflectance map 60 of the buttocks 84 generated by the analysis means 30 and determine the boundary position between the heartwood 92 and the sapwood 88. Furthermore, the artificial intelligence unit 38 is configured to perform additional learning to determine the boundary position between the heartwood 92 and the sapwood 88, using the infrared reflectance map 60 of the buttocks 84 stored in the storage means 50 (see S100 in Fig. 2). This allows the artificial intelligence unit 38 to more accurately distinguish between the heartwood 92 and the sapwood 88, and even if the boundary position between the heartwood 92 and the sapwood 88 has an irregular shape, the artificial intelligence unit 38, which performs additional learning, can determine the boundary position without any problems.

[0048] On the other hand, the log analysis method according to the embodiment of the present invention can be executed using the log analysis system 10 according to the embodiment of the present invention as described above, thereby achieving the same effects as those of the log analysis system 10 according to the embodiment of the present invention. [Explanation of symbols]

[0049] 10: Log analysis system, 14: Photography means, 18: Depth camera, 22: Infrared measurement means, 30: Analysis means, 38: Artificial intelligence unit, 50: Storage means, 60: Reflectance map, 80: Log, 84: End grain, 88: Sapwood, 92: Heartwood

Claims

1. A photographing means for photographing the end of a log; An infrared measuring means for measuring the reflectance of infrared rays from the end of the wood; an analysis means for generating an infrared reflectance map of the butt end from the image taken by the photographing means and the measurement results of the infrared measuring means, and for determining the boundary position between the heartwood and sapwood at the butt end based on the reflectance map.

2. The photographing means includes a depth camera that photographs the end grain and measures the depth of an object in the photographed image, The log analysis system according to claim 1, characterized in that the analysis means determines the contour of the end grain in the captured image from the captured image and depth information obtained from the depth camera, and specifies the range for generating the reflectance map using the determined contour of the end grain.

3. 3. The log analysis system according to claim 2, wherein said analysis means uses relative infrared reflectance within said specified range when generating said reflectance map.

4. The log analysis system according to claim 2, wherein the analysis means determines vertical and horizontal end points of the end point in the photographed image from the determined contour of the end point, and calculates the vertical and horizontal widths of the end point based on the depth information of each determined end point.

5. 5. The log analysis system according to claim 4, wherein the analysis means calculates the sizes of the heartwood and the sapwood based on the determined boundary position, the determined contour of the butt end, and the depth information.

6. 6. A log analysis system according to claim 5, wherein said analysis means outputs an instruction for a cutting position of said log based on the content of the analysis.

7. further comprising a storage means for storing analysis results of the analysis means, including the reflectance map; the analyzing means includes an artificial intelligence unit for analyzing the reflectance map to determine the boundary location; 2. The log analysis system according to claim 1, wherein the artificial intelligence unit performs additional learning using the reflectance map stored in the storage means.

8. Photographing the end of the log, Measure the reflectance of infrared light from the end grain, A log analysis method characterized by generating an infrared reflectance map of the butt end from the photographed image of the butt end and the measurement results of the infrared reflectance, and determining the boundary position between the heartwood and sapwood at the butt end based on the reflectance map.

9. When photographing the end grain, the depth of an object in the photographed image is measured, The log analysis method according to claim 8, characterized in that the contour of the end grain in the photographed image is determined from the photographed image and the measured depth information, and the determined contour of the end grain is used to specify the range for generating the reflectance map.

Images