Apparatus and program for evaluating mold surface roughness

Abstract

To provide an apparatus and a program that can accurately evaluate mold surface roughness from image data.SOLUTION: An apparatus for evaluating mold surface roughness includes: an acquisition unit that acquires image data to which a mold surface is mapped; a threshold setting unit that determines a threshold based on the standard deviation of the lightness of pixels that make up the image data; a moving average calculation unit that calculates a moving average of the lightness of pixel data; a void determination unit that determines a pixel where a difference between the lightness of the pixel data and the moving average of the pixel is greater than the threshold to be a void; and a void ratio calculation unit that calculates a void ratio defined as a ratio of voids to the pixels that make up the image.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] The present invention relates to an apparatus and a program for evaluating the surface roughness of a mold, particularly a mold made of green sand. [Background technology]

[0002] The surface roughness of a mold not only affects the surface roughness of the resulting casting, but also correlates strongly with the mold's density. Furthermore, mold density correlates strongly with mold strength, which in turn influences the dimensional accuracy of the resulting casting. Therefore, accurately assessing the mold's surface roughness before casting is crucial for stabilizing the quality of sand-cast castings. For this reason, for example, in the operation of a mass-production line for green sand castings, workers sometimes measure mold strength by pressing a dedicated mold strength meter directly against the mold surface to prevent casting into a defective mold or to predict molding machine malfunctions.

[0003] However, measurements using mold strength meters destroy part of the mold, making it impossible to measure the cavity surface, which directly affects the quality of the casting. For this reason, technologies have been proposed for non-contact measurement of mold density, which correlates with mold strength. For example, Patent Document 1 discloses a "method for inspecting the density of a mold made of refractory particles, comprising: irradiating a laser beam onto the mold surface with a laser emitter; observing the laser spot on the mold surface with a receiver; simultaneously moving the laser emitter and receiver together a predetermined distance along the mold surface; and counting the number of times the receiver fails to observe the laser spot during the predetermined distance movement, thereby inspecting the mold density." This technology utilizes the fact that the surface irregularities of a sand mold have a strong correlation with mold density. It scans the mold surface with a laser beam and evaluates the irregularities to assess the mold density. However, scanning the required area with a laser beam is expected to take a considerable amount of time.

[0004] In response to this, in recent years, devices have been proposed that inspect the appearance of an object using captured images. Compared to the above-mentioned laser scanning operation, image acquisition time can be significantly reduced, and the captured image data can be processed at high speed by a computer, thereby significantly reducing inspection time. For example, Patent Document 2 discloses "an apparatus for inspecting the appearance of an object, comprising: an imaging device that images the object from a first direction; an illumination unit that irradiates the object with light using a first irradiation pattern that irradiates light from a first position and a second irradiation pattern that irradiates light from a second position different from the first position; and a controller that acquires a first inspection image by causing the imaging device to image the object irradiated with light using the first irradiation pattern, acquires a second inspection image by causing the imaging device to image the object irradiated with light using the second irradiation pattern, and inspects the appearance of the object based on the first inspection image, the second inspection image, and a predetermined reference image, wherein the first position and the second position overlap each other when viewed from the first direction." This technology improves inspection accuracy by arranging light sources that illuminate the object in two different directions, reducing the areas of the object's surface that are not illuminated by light, thereby enabling the acquisition of high-quality image data.However, in foundries where there is generally a lot of dust and it is difficult to ensure a good shooting environment, if this technology is to be applied to measuring the surface roughness of molds, additional equipment such as dust countermeasures to stabilize the irradiated light will likely be required.

[0005] On the other hand, even when image data is captured under such constraints in the imaging environment, a technology has been proposed that clarifies desired portions of an image through binarization processing. For example, Patent Document 3 discloses an image binarization method for binarizing a two-dimensional image with a grayscale dot configuration, such as one that reflects the shape of an object using reflected light, characterized by creating a brightness histogram for each vertical row of dots in the image, and performing binarization processing by setting the brightness level of a predetermined number of dots, counting from the highest brightness side, to a predetermined high brightness and the brightness level of the remaining dots to a predetermined low brightness. This method is described as "an image binarization method that clearly represents the shape when binarizing an image with linear contours and no vertical overlapping, such as one that reflects the shape of an object using reflected light, and that requires a short processing time, small memory capacity, and an inexpensive processing device," but is not considered suitable for evaluating the surface roughness of a mold. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Japanese Patent Application Publication No. 7-132345 [Patent Document 2] International Publication No. 2018 / 216495 [Patent Document 3] Japanese Patent Application Publication No. 7-065158 Summary of the Invention [Problem to be solved by the invention]

[0007] When attempting to accurately evaluate the surface roughness of a mold, particularly one made from green sand, solely by analyzing image data of the mold surface, it is desirable that the entire image be captured with uniform exposure. However, foundries generally do not provide an environment with sufficient exposure, resulting in undulations in the image. For this reason, it has been difficult to accurately evaluate the mold surface roughness using conventional processing methods for image data captured in an environment where sufficient exposure is not available. In response to these challenges, the present invention aims to provide an apparatus and program for evaluating the surface roughness of a mold from image data. [Means for solving the problem]

[0008] The first embodiment of the present invention as a means for solving the above problems is as follows: An apparatus for evaluating the surface roughness of a mold, comprising: an acquisition unit that acquires image data of the surface of the mold; a threshold setting unit that determines a threshold based on a standard deviation of brightness of pixels that constitute the image data; a moving average calculation unit that calculates a moving average of the brightness of the pixel data; a gap determination unit that determines a pixel, the difference between the brightness of the pixel data and the moving average of the pixel being greater than the threshold, to be a gap; a void ratio calculation unit that calculates a void ratio defined as a ratio of the voids to the pixels that constitute the image; The apparatus for evaluating the surface roughness of the mold comprises:

[0009] A preferred embodiment of the device is further comprising an image division unit that divides the image data into a plurality of mutually different parts; the threshold setting unit determines the threshold for each of the body parts, the moving average calculation unit calculates the moving average for each of the parts, the gap determination unit determines the gap for each of the portions, The porosity calculation unit calculates the porosity for each of the portions. It is a device.

[0010] A further preferred embodiment of the device is The apparatus further comprises a statistical processing unit that statistically processes the porosity for each of the portions to obtain a statistical value.

[0011] Moreover, the second embodiment of the present invention is Acquire image data of the surface of the mold; determining a threshold value based on a standard deviation of brightness of pixels constituting the image data; Calculating a moving average of the brightness of the pixel; determining a pixel having a difference between the brightness of the pixel and the moving average of the pixel that is greater than the threshold value to be a void; calculating a porosity defined as the ratio of the voids to the pixels constituting the image; It is a program that causes a computer to execute the above.

[0012] A preferred form of the program is: Dividing the acquired image into a plurality of different regions; determining the threshold value for each of the regions; The moving average is calculated for each of the parts; determining the voids for each of the portions; determining the porosity for each of the portions; It is a program that causes a computer to execute the above.

[0013] A more preferred embodiment of the program is Statistically processing the porosity determined for each of the portions to determine a statistical value. It is a program that causes a computer to execute the above. [Effects of the Invention]

[0014] The present invention makes it possible to accurately measure the surface roughness of a mold even from images captured in an environment where sufficient exposure is not possible. This makes it possible to estimate mold density, which has a strong correlation with surface roughness, in-line, and stabilize casting production by identifying abnormal conditions in the molding machine in-line and eliminating defective molds in advance. [Brief explanation of the drawings]

[0015] [Figure 1] 1 is a schematic diagram showing an example of the configuration of a system including an apparatus according to the present invention. [Figure 2] 1 is a schematic diagram showing an example of the configuration of an apparatus according to the present invention. [Figure 3] FIG. 2 is a schematic diagram showing an example of the function of the device according to the present invention. [Figure 4] 1 is a flow chart showing an example of a procedure for determining void ratio using the device according to the present invention. [Figure 5] 1 is a flow chart showing an example of a method for evaluating the surface roughness of a mold according to the present invention. [Figure 6] FIG. 3 is a diagram showing an example of a brightness profile in image data according to the first embodiment of the present invention. [Figure 7] 1 is a graph showing an example of the relationship between porosity and mold density determined according to the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0016] First, referring to Figure 1, we will explain an example of a system 200 (hereinafter also referred to as an evaluation system) for evaluating the surface roughness of a mold, which includes an apparatus 1 (hereinafter also referred to as an evaluation apparatus) for evaluating the surface roughness of a mold according to a first embodiment of the present invention.

[0017] The evaluation system 200 includes an evaluation device 1. The evaluation device 1 is connected to a photographing device 220 via a telecommunications line 210 and may also be connected to a server 230 or the like. The photographing device 220 is a device for photographing the surface of the mold 300, which is the subject of the present invention, and includes at least a camera 221 for photographing the surface of the mold 300, a robot arm 222 for holding the camera 221 and determining an arbitrary photographing position on the surface of the mold 300, and a terminal 223 for controlling the operation of the robot arm 222 and the camera 221. The image data photographed by the photographing device 220 may be transmitted directly to the evaluation device 1 via the telecommunications line 210, or may be transmitted to the server 230 and stored as a predetermined number of image data. The evaluation device 1 obtains image data of the surface of the mold 300 from the photographing device 220 or the server 230. 1 shows a configuration in which the evaluation device 1 acquires image data via the telecommunications line 210, but the evaluation device 1 may also be configured to acquire image data directly without via the telecommunications line 210 by further providing the functionality of the photographing device 220. In this way, photographed images are generally acquired or saved in the form of digital data, and visualized images are one form of this. Therefore, both digital and visualized formats are referred to as image data here (the same applies hereinafter).

[0018] Next, an example of the evaluation device 1 according to the first embodiment of the present invention will be described. Fig. 2 is a schematic diagram showing an example of the configuration of the evaluation device 1, and Fig. 3 is a schematic diagram showing an example of the function of the evaluation device 1.

[0019] See Figure 2. The evaluation device 1 is composed of a main body 10, which is its main part, and an input device 12 and a display device 13, which are directly connected to the main body 10. The evaluation device 1 is specifically a computer, and in addition to electronic devices such as a workstation (WS) or a personal computer (PC), electronic devices such as smartphones, tablet terminals, wearable terminals, IoT (Internet of Things) devices, and single-board computers such as Raspberry Pi (registered trademark) may also be used, and these terminals may have built-in microphones, cameras, etc.

[0020] 2, the main body 10 of the evaluation device 1 includes, for example, a CPU (Central Processing Unit) 101, a RAM (Random Access Memory) 102, a ROM (Read Only Memory) 103, an STR (Storage) 104, which is a storage device such as an HDD (Hard Disk Drive) or an SSD (Solid State Drive), and an I / F (Interface) 105, and may further include a GPU (Graphics Processing Unit, a processor dedicated to image processing) 106. These components 101 to 106 are connected to each other by an internal bus 107. The main body 10 may further include a housing 100 that houses these components.

[0021] The CPU 101 controls the entire evaluation device 1. The RAM 102 is a work area used when the CPU 101 is operating. The ROM 103 stores the operation code of the CPU 101. If a GPU 106 is provided, it processes data converted and processed by the CPU 101 in parallel, contributing to high-speed calculation processing. The STR 104 is an auxiliary storage area for recording or reading information controlled or processed by the CPU 101. The I / F 105 is an interface for transmitting and receiving various information to and from external devices such as the input device 12 and display device 13 connected to the main body 10.

[0022] Of these I / Fs 105, the I / F 108 transmits and receives image data and various information to and from an imaging device 220, a server 230, etc. via a telecommunications line 210 in the evaluation system 200 shown in FIG.

[0023] The I / F 109 transmits and receives information to and from the input device 12. A plurality of input devices 12 may be provided, and may include, for example, a command or numerical value input means such as a keyboard, an image acquisition means such as an image scanner, a pointing device such as a mouse, etc. An operator or the like who uses the evaluation device 1 inputs commands and various information for controlling the evaluation device 1 via these input devices 12.

[0024] Furthermore, the I / F 110 transmits and receives various types of information to and from the display device 13. The display device 13 outputs various types of information stored in the storage device STR 104, the processing status of the evaluation device 1, and the like. The display device 13 may be, for example, a display. The display may also function as the input device 12, such as a touch panel type. Furthermore, a plurality of display devices 13 may be provided.

[0025] Various information including image data input to the main body 10 of the evaluation device 1 may be input directly by an operator or the like via the input device 12, or may be input as a signal transmitted from an imaging device 220 or a server 230 connected via a telecommunications line 210 in the evaluation system 200 shown in Fig. 1. This various information may also include information for operating or controlling the evaluation device 1.

[0026] Next, the function of the evaluation device 1 will be described with reference to FIG. (1) Acquisition section 151 The acquisition unit 151 acquires image data of the surface of the mold 300 (see FIG. 1). The image data may be a color image containing color information, or a grayscale image containing only brightness information (hereinafter also referred to as brightness) and no color information. It is preferable that the acquisition unit 151 also has a function to acquire all information (hereinafter referred to as adjustment information) required by the operator to operate the evaluation device 1 under desired conditions, such as information for adjusting the brightness of the pixels constituting the image data, information for setting a threshold in the threshold setting unit 152 (described below), information for calculating a moving average in the moving average calculation unit 153 (described below), information for providing conditions for determining voids in the void determination unit 154 (described below), and information for segmenting the image data into multiple regions in the image segmentation unit 156. This allows the operator of the evaluation device 1 to provide this adjustment information, enabling the adjustment unit 159 and statistical processing unit 160 (described below) to obtain results reflecting the desired adjustments and statistical values.

[0027] (2) Threshold setting unit 152 The threshold setting unit 152 determines a threshold based on the standard deviation of the brightness of the pixels constituting the image data acquired by the acquisition unit 151. Here, the "brightness" handled by the threshold setting unit 152 will be explained. For example, a grayscale image is generally visualized as a black-and-white image with 256 levels ranging from 0 (black) to 255 (white). These 256 integer levels represent brightness. While not only grayscale images but also color images contain such brightness information, the threshold setting unit 152 handles only the brightness information contained in the image data, regardless of the presence or absence of color information. By handling only brightness in this way, the processing steps for calculating the porosity (described later) can be simplified and processed at high speed. Note that brightness values are not limited to integer values of 256 levels, and any level from black (no brightness) to white (maximum brightness, no blackness) may be expressed numerically. For example, 16 levels or 1024 levels may be used, or values obtained by converting these integer values to numerical values from 0 to 100 may be used.

[0028] The threshold setting unit 152 then has the function of calculating the standard deviation σ of the brightness of the pixels that make up the image data and setting the threshold TH based on this standard deviation σ. Here, "based on the standard deviation σ" means that the only variable in the calculation formula that determines the threshold TH is the standard deviation σ. For example, this refers to a case where the threshold TH is defined as a linear expression of the standard deviation σ, such as TH=aσ+b, where a and b are predetermined constants (except a=0). In this linear expression example, the predetermined constants a and b may be arbitrarily defined in advance. For example, a=0.5, b=0, and a value 0.5 times the standard deviation may be used as the threshold. Furthermore, the expression that defines the threshold is not limited to such a linear expression, and may be any function as long as the standard deviation is the only variable.

[0029] (3) Moving average calculation unit 153 The moving average calculation unit 153 calculates a moving average of the brightness of pixels constituting the image data acquired by the acquisition unit 151. Generally, a moving average in image processing technology is a filtering technique used to reduce image noise. However, the moving average in the present invention is not used to reduce image noise, but is calculated as a parameter for identifying pixels corresponding to voids (described later). In image processing, the moving average refers to the average brightness of the pixels surrounding the pixel in question (window size) including the pixel itself. In the present invention, it is preferable to consider a predetermined number of pixels in a one-dimensional range adjacent to the pixel in either the row or column to which the pixel belongs, rather than the pixels surrounding the pixel in question (a two-dimensionally expanded range). This is because, since incident light from a certain direction tends to be strong in foundries, using a one-dimensional moving average may be more reliable than a two-dimensional moving average for image data captured in such an environment in determining the presence of voids.

[0030] Therefore, it is preferable that the moving average calculation unit 153 of the present invention obtains the brightness values of the pixel of interest (hereinafter also referred to as the pixel in question) itself and a predetermined number of pixels adjacent to the pixel in question in either the row or column to which the pixel belongs, and uses the average of these values as the moving average of the pixel in question. For example, the sum of the brightness values of 101 (a predetermined number) pixels, including the pixel in question itself and 50 pixels on each side adjacent to the pixel in the same row, is divided by the predetermined number, 101, and the quotient is used as the moving average of the pixel in question. The predetermined number may be set arbitrarily. Using this method, the moving average calculation unit 153 calculates the moving average of all pixels that make up the acquired image data.

[0031] (4) Gap determination unit 154 The void determination unit 154 determines a pixel as a void if the difference between the brightness of each pixel constituting the image data and the moving average of each pixel calculated by the moving average calculation unit 153 is greater than the threshold value set by the threshold setting unit 152. Although the surface of an actual mold may be smooth macroscopically, irregularities exist microscopically. In this invention, relatively recessed portions are referred to as voids. When the surface of a mold is observed under visible light, recessed portions appear darker than protruding portions due to the reduced amount of reflected light. Therefore, microscopically dark portions are likely to be voids. However, in an insufficient exposure environment, such as in a foundry, it is difficult to capture the surface of a mold with uniform exposure, resulting in variations in brightness within the same image, i.e., portions that appear excessively bright or excessively dark. Therefore, when identifying voids in a mapped image, in order to smooth out fluctuations in brightness and accurately reflect the true number and distribution of voids, the condition for determining a void in the present invention is that the difference between the brightness of each mapped pixel and the moving average of that pixel is greater than a set threshold value.

[0032] (5) Porosity calculation section 155 The porosity calculation unit 155 calculates the porosity defined as the ratio of the voids determined by the void determination unit 154 to the pixels constituting the image data. That is, the ratio of the number of pixels determined to be voids to the total number of pixels constituting the image data is calculated as the porosity of the image data.

[0033] (6) Image division unit 156 The image division unit 156 divides the image data acquired by the acquisition unit 151 into different regions. Image data captured in an environment where incident light from a specific direction is strong tends to have more pronounced fluctuations in brightness as the captured range (area) becomes larger. For this reason, when the number of pixels in the image data is large, the image data is divided into pixel groups of different predetermined ranges, a threshold is set for each divided region, a moving average is calculated, voids are determined, and the porosity is calculated. This smooths out the fluctuations in brightness for each region, allowing for a more accurate porosity. For this reason, it is preferable to have the image division unit 156. Information for division is provided by the adjustment unit 159, which will be described later.

[0034] (7) Output section 157 The output unit 157 outputs the porosity calculated by the porosity calculation unit 155. The output unit 157 may also have a function to further output the image data acquired by the acquisition unit 151, the standard deviation calculated by the threshold setting unit 152, the threshold and a calculation formula defining the threshold, the moving average calculated by the moving average calculation unit 153, the pixels and the number of pixels for calculating the moving average, and adjustment information such as information about the sections divided by the image division unit 156, and statistical values of the porosity calculated by the statistical processing unit 160 described later, and may also have a function to appropriately output the processing status of the evaluation device 1.

[0035] (8) Storage section 158 The memory unit 158 has the function of storing and reading out various information processed by the acquisition unit 151, threshold setting unit 152, moving average calculation unit 153, void determination unit 154, void ratio calculation unit 155, image division unit 156, output unit 157, and the adjustment unit 159 and statistical processing unit 160 described later in RAM 102 and STR 104 (see Figure 2).

[0036] (9) Adjustment section 159 The evaluation device 1 may also include an adjustment unit 159. The adjustment unit 159 has a function of adjusting the operations of the threshold setting unit 152, the moving average calculation unit 153, the void determination unit 154, the void ratio calculation unit 155, the image division unit 156, the output unit 157, the storage unit 158, and the statistical processing unit 160, which will be described later, based on the adjustment information acquired by the acquisition unit 151.

[0037] (10) Statistical processing unit 160 Furthermore, the evaluation device 1 may have a statistical processing unit 160. The statistical processing unit 160 has a function of statistically processing the porosity calculated for each region divided by the image division unit 156 and calculating a statistical value of the porosity.

[0038] Next, a second embodiment of the present invention, a program (hereinafter also referred to as this program) that causes a computer to execute to determine the porosity, which is defined as the proportion of voids in the pixels that make up an image of the surface of a mold, will be described with reference to the flowchart shown in FIG. 4, as well as FIG. 1 showing evaluation system 200 and FIG. 3 showing the functions of evaluation device 1.

[0039] (1) Step of acquiring image data (S1) In the step (S1) of acquiring image data (hereinafter also referred to as the S1 step, and the same applies to the other steps below), image data of the surface of the mold 300 imaged using the photographing device 220 shown in Figure 1 is acquired from the photographing device 220 (terminal 223) or the server 230 by operating the acquisition unit 151 (see Figure 3) of the evaluation device 1.

[0040] (2) Step S2: Determining a Threshold Next, the step (S2) of determining a threshold value based on the standard deviation of the brightness of the pixels that make up the image data acquired in step S1 is executed by operating the threshold value setting unit 152 (see FIG. 3).

[0041] (3) Step S3: Calculating the moving average Furthermore, the step (S3) of calculating the moving average of the brightness of the pixels that make up the image data acquired in step S1 is executed by operating the moving average calculation unit 153 (see FIG. 3).

[0042] (4) Step (S4) of determining whether it is a void Next, based on the threshold value determined in step S2 and the moving average calculated in step S3, step S4 is executed by operating the gap determination unit 154 (see Figure 3) to determine that pixels constituting the image data whose difference between the brightness of the pixel and the moving average of the pixel is greater than the threshold value are gaps.

[0043] (5) Step of calculating porosity (S5) Then, the porosity calculation unit 155 (see Figure 3) is operated to execute a step (S5) of calculating the porosity, which is defined as the proportion of pixels determined to be voids in step S4 to the pixels that make up the entire image data.

[0044] Furthermore, the program may be configured to execute steps such as, as needed, operating the output unit 157 to output the porosity obtained in step S5 to the output device 13 (not shown), operating the memory unit 158 to store the porosity in the STR 104 (not shown), or operating the adjustment unit 159 to adjust the porosity based on adjustment information input via the input device 12 (not shown).

[0045] The program may also be configured to execute a step (not shown) of dividing the image data acquired in step S1 by operating the image division unit 156, and to execute steps S2 to S5 for each divided area.

[0046] Next, a method for evaluating the surface roughness of a mold (hereinafter also referred to as the present method), which is yet another embodiment of the present invention, will be described with reference to the flowchart shown in FIG.

[0047] (1) Step (s1) of preparing image data First, image data onto which the surface of the mold is mapped is prepared (s1). The form of the prepared image data may be a form photographed by the photographing device 220 as shown in Fig. 1, or a form stored in advance in the server 230, or may be in the form of image data photographed separately. Furthermore, the prepared image data may be input to the evaluation device 1 via the telecommunications line 210 (see Fig. 1) or via the input device 12 (see Fig. 2).

[0048] (2) Step (s2) of determining a threshold value A threshold is determined based on the standard deviation of the brightness of the pixels that make up the image data prepared in step s1 (s2). The method for determining the threshold may be the same as the method described above in the explanation of threshold setting unit 152.

[0049] (3) Step s3: Calculating the moving average In conjunction with step s2, a moving average of the brightness of the pixels constituting the image data prepared in step s1 is calculated (s3). The method for calculating the moving average may be the same as the method described above in the description of the moving average calculation unit 153.

[0050] (4) Step (s4) of determining whether it is a void Based on the threshold value determined in step s2 and the moving average calculated in step s3, pixels for which the difference between the brightness of each pixel constituting the image data and the moving average of that pixel is greater than the threshold value are determined to be voids (s4). The method for determining voids may be the same as the method described above in the description of void determination unit 154.

[0051] (5) Step (s5) for calculating porosity The ratio of the number of voids (pixels) determined in step s4 to the number of all pixels constituting the image data is calculated as the void ratio (s5).

[0052] (6) Step (s6) of evaluating surface roughness The surface roughness of the region where the mold is imaged is evaluated based on the porosity determined in step s5 (s6). That is, the porosity determined in step s5 is used as an index of the surface roughness of the region where the mold is imaged.

[0053] Step s1 of this method may further include a step of dividing the prepared image data into a plurality of mutually different regions (not shown, hereinafter also referred to as the dividing step). In particular, when the number of pixels in the image data is large, after the dividing step, a step (s2) of determining a threshold, a step (s3) of calculating a moving average, a step (s4) of determining voids, and a step (s5) of calculating the void ratio are performed for each divided region, thereby enabling a more accurate evaluation of the void ratio, i.e., a more accurate surface roughness, in which the undulations in brightness are leveled for each region (s6).

[0054] As described above, when evaluating the surface roughness of a mold based on the surface roughness of each region, the divided regions are represented by multiple surface roughness (porosity) values. However, particularly when there are many divided regions, it may be more convenient to statistically process these multiple surface roughness (porosity) values and convert them into a representative value (statistical value). In such cases, step s5 may further include a step (not shown) of statistically processing the porosity of each region to obtain a statistical value, and the surface roughness of the mold is evaluated based on this statistical value. Here, the statistical value refers to one or more values obtained by statistically processing the porosity of each region. Examples include, but are not limited to, the mean, median, and percentile value. This allows for the acquisition of a surface roughness statistical value (representative value) that reflects the entire image data when the porosity of multiple regions is present in a single image data. For example, for molds with a large mapping area, the representative surface roughness value (statistical value) can be used to easily compare differences in surface roughness between molds and to understand changes in the operating status of the molding machine over time. [Example]

[0055] The following examples of the present invention will be described. However, these examples are merely examples of the present invention and the present invention is not limited to these examples.

[0056] The mold to be photographed was one made by tamping green sand with silica sand as the aggregate, and the surface of the mold to be photographed was a flat circular area with a diameter of approximately 50 mm. The camera used was a digital camera (OLYMPUS E-M10II) with 17 million effective pixels and a macro lens (ZUIKO-DIGITAL) with a focal length of 35 mm attached. The photographing range was adjusted to include the aforementioned circular area, and images were taken from a direction perpendicular to the surface of the mold, obtaining image data which was then input into evaluation device 1 (PC).

[0057] Below, a program written in the programming language Python (registered trademark) was executed, causing the evaluation device 1 to perform the steps of classification, setting a threshold, calculating a moving average, determining whether it is a void, calculating the void ratio, and calculating statistical values.

[0058] Here, the brightness information of the image data, which is an integer value from 0 to 255, was converted to a value from 0 to 100, with the maximum value of 255 being 100, and this value was used as the brightness. The mapping area of the image data was divided into 100 distinct regions (pixel groups) (= 10 vertical x 10 horizontal), the threshold value TH was set to 0.5 times the standard deviation σ (TH = 0.5σ), and the predetermined number of pixels used to calculate the moving average was set to 100 pixels, 50 pixels before and 50 pixels after the pixel belonging to the row of the pixel in question, which were input into the evaluation device 1 as adjustment information, to calculate a moving average (100-point moving average) that also included the brightness of the pixel itself. Based on the threshold value and moving average set as above, pixels where the difference between the moving average of the pixel and the brightness of the pixel in question was greater than the threshold were determined to be voids.

[0059] FIG. 6 is a diagram showing an example of a brightness profile in image data in this embodiment. A portion of image data 910 in FIG. 6, used for explanation, is a portion of image data acquired in this embodiment and is visualized as a grayscale image consisting only of brightness information. The horizontal axis of graph 920 in FIG. 6 indicates the positions of scanning target pixels 911 belonging to the row indicated by the white dashed line in the portion of image data 910, numbered in ascending order starting from 1, when scanning in a scanning direction 912 (the direction of the right-pointing arrow). Meanwhile, the vertical axis indicates values obtained by converting the brightness range of the pixel, which ranges from 0 to 255, into a range from 0 to 100. Brightness 921, indicated by a solid line in graph 920, indicates the brightness of the scanning target pixel at each position indicated by the horizontal axis. Furthermore, average value 922, indicated by a dashed-dotted line, indicates the average brightness (value 55) of all pixels constituting the divided region (pixel group) to which the portion of image data 910 belongs, and is a value used together with brightness 921 to determine the threshold. The standard deviation of brightness for the portion (pixel group) to which the image data of this embodiment belongs was found to be 25.6 (not shown), and the threshold, which is 0.5 times this standard deviation, was found to be 12.8 (not shown). The moving average 923 indicated by the two-dot chain line represents the moving average of the scanning target pixel 911. In graph 920, the difference between the moving average and brightness at scanning target pixel positions 1 to 50 was greater than the threshold, and the positions of the pixels determined to be voids were seven pixels, from 36 to 42 (the range 913 of pixels determined to be voids, indicated by the double-arrowed line in image data 910).

[0060] The porosity of each of the 100 sections was determined using the above method. The average porosity of the 40 sections with the highest porosity was then calculated as a statistical value. This statistical value was then redefined as the porosity representative of the entire image data and used as an index of the surface roughness of the mold in the corresponding mapping region. The porosity (statistical value) in this example was 16.8%.

[0061] Figure 7 is a graph showing an example of the relationship between void fraction and mold density. The vertical axis represents the density index (unitless, absolute number) which is a relative value based on the density of the mold (sand mold) of the example shown in Figure 6. The horizontal axis represents the void fraction (statistical value, the same applies below) calculated in the same manner as in the above example. Point A in Figure 7 plots the density index (1.000) at the void fraction (16.8%) of the mold (sand mold) used in the above example. Points B and C show the relationship between the density index and void fraction, calculated using the same method as in the above example from image data of the same region as in the above example for two sand molds of known different densities. Point B had a void fraction of 17.8% and a density index of 0.990, and point C had a void fraction of 20.4% and a density index of 0.973. These three points demonstrate the linear relationship shown by the dashed line in Figure 7, demonstrating that the density of a sand mold can be accurately predicted by calculating the void fraction according to the present invention. [Explanation of symbols]

[0062] 1: Evaluation device 10: Main body 100: Housing 101:CPU 102:RAM 103:ROM 104: Storage (STR) 105(108,109,110):I / F 106: GPU 107: Internal bus 151: Acquisition Department 152: Threshold setting unit 153: Moving average calculation section 154:Gap determination section 155: Porosity calculation part 156: Image division section 157: Output section 158: Storage section 159: Adjustment section 160: Statistical processing unit 12: Input device 13: Display 200: Rating System 210: Telecommunications lines 220: Imaging device 221: Camera 222:Robot arm 223: Terminal 230: Server 300: Mold 910: Part of image data 911: Scanning target pixel 912: Scanning direction 913: Range of pixels determined to be void 920:Graph 921: Brightness 922:Average value 923: Moving average

Claims

1. An apparatus for evaluating the surface roughness of a mold, comprising: an acquisition unit that acquires image data of the surface of the mold; a threshold setting unit that determines a threshold based on a standard deviation of brightness of pixels that constitute the image data; a moving average calculation unit that calculates a moving average of the brightness of the pixel data; a gap determination unit that determines a pixel, the difference between the brightness of the pixel data and the moving average of the pixel being greater than the threshold, to be a gap; a void ratio calculation unit that calculates a void ratio defined as a ratio of the voids to the pixels that constitute the image; An apparatus for evaluating the surface roughness of a mold, comprising:

2. further comprising an image division unit that divides the image data into a plurality of mutually different parts; the threshold setting unit determines the threshold for each of the body parts, the moving average calculation unit calculates the moving average for each of the parts, the gap determination unit determines the gap for each of the portions, The porosity calculation unit calculates the porosity for each of the portions.

10. The apparatus of claim 1.

3. The apparatus according to claim 2 , further comprising a statistical processing unit that statistically processes the porosity for each of the regions to obtain a statistical value.

4. Acquire image data of the surface of the mold; determining a threshold value based on a standard deviation of brightness of pixels constituting the image data; Calculating a moving average of the brightness of the pixel; determining a pixel having a difference between the brightness of the pixel and the moving average of the pixel that is greater than the threshold value to be a void; calculating a porosity defined as the ratio of the voids to the pixels constituting the image; A program characterized by causing a computer to execute the above.

5. Dividing the acquired image data into a plurality of different regions; determining the threshold value for each of the regions; The moving average is calculated for each of the parts; determining the voids for each of the portions; determining the porosity for each of the portions; The program according to claim 4, which causes a computer to execute the steps.

6. Statistically processing the porosity determined for each of the portions to determine a statistical value. The program according to claim 5, which causes a computer to execute the steps.

Images