Sand mold individual identification system, sand mold data input system, and sand mold data verification system

Abstract

To provide a sand mold individual identification system that can identify individual sand molds, with equipment costs being inhibited.SOLUTION: The individual sand mold individual identification system includes: a surface photographing unit 13 that photographs a surface pattern of a sand mold; a shape model database 15 that stores a first shape model obtained from photograph data for each sand mold photographed by the surface photographing unit 13 in association with each manufacturing condition; a surface photographing unit 23 that photographs the surface pattern of the sand mold in a use stage during casting; a shape model collating unit 25 that identifies the first shape model that matches a second shape model obtained from photograph data of the surface photographing unit 23; and a sand mold database 26 that stores the manufacturing condition stored in the shape model database 15 corresponding to the first shape model identified by the shape model collating unit 25 in association with an identification ID assigned to a cast product manufactured using the sand mold.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to a mold individual identification system for identifying an individual mold used in casting, and a mold data input system and a mold data verification system that constitute the mold individual identification system.

Background Art

[0002] In industrial products, various parts are produced by casting. For example, in an engine, a cylinder block and a cylinder head are produced as castings. Further, when a casting is a part having a cavity inside, a main mold (main pattern) and a core are used as molds for casting. The main mold is a mold that forms the outside of the casting, and the core is a mold that forms the cavity portion of the casting. The mold used in casting is destroyed at the stage of taking out the casting from the mold.

[0003] Castings produced by sand casting generally have a high defect rate (about several percent). Further, the influence of the quality of the mold on the casting is great. Therefore, in order to reduce the defect rate of castings and improve the yield, it is preferable to be able to trace the data of the mold (manufacturing conditions of the mold, etc.) used at the time of casting from the final casting (to be able to individually identify the mold). This makes it easier to identify the cause in the mold when a casting defect occurs.

[0004] Patent Document 1 discloses that a management code is engraved on a mold by laser irradiation or the like, and the mold is individually identified by this management code. Since the management code engraved on the mold is transferred to the casting, the data of the mold can be traced from the management code even after casting (after the mold is destroyed).

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0006] When a management code is engraved on a sand mold, an engraving device such as a laser irradiation device is required, resulting in an increase in equipment costs. In particular, when trying to install an engraving device for individual identification in the production line for sand molds (e.g., cores) that have not been individually identified so far, a line change is also required, incurring costs for that.

[0007] The present invention has been made in view of the above problems, and an object thereof is to provide a sand mold individual identification system capable of performing individual identification of sand molds while suppressing equipment costs.

Means for Solving the Problems

[0008] In order to solve the above problems, a sand mold individual identification system according to a first aspect of the present invention includes a first imaging unit that images a pattern on the surface of a sand mold, a first database that stores, for each sand mold imaged by the first imaging unit, a first shape model obtained from imaging data in association with the manufacturing conditions of the sand mold, a second imaging unit that images a pattern on the surface of the sand mold used during casting, a collation unit that identifies the first shape model that matches the second shape model obtained from the imaging data of the second imaging unit, and a second database that stores, in association with the identification ID assigned to the casting product manufactured using the sand mold, the manufacturing conditions recorded in the first database corresponding to the first shape model identified by the collation unit.

[0009] According to the above configuration, individual identification can be performed by using the random pattern (shape model of sand grains) that naturally occurs on the surface of the sand mold without engraving a management code or the like on the sand mold. That is, by comparing the second shape model obtained by photographing with the second photographing unit with the first shape model obtained by photographing with the first photographing unit in advance, the manufacturing conditions of the sand mold associated with the first shape model in the sand mold manufacturing stage can be associated with and stored as the identification ID given to the casting product manufactured using the sand mold in the casting stage. At this time, an engraving device such as a laser irradiation device is not required in the sand mold manufacturing line, and the equipment cost can be suppressed.

[0010] Further, in the above sand mold individual identification system, the first shape model or the second shape model can be configured to be obtained by extracting the edge shape of the sand grain pattern in the photographing data by the first photographing unit or the second photographing unit through image processing.

[0011] According to the above configuration, since the edge-extracted shape model uses the shape of the sand grains, it becomes easier to recognize even if there are color changes or dirt on the surface of the sand mold during the process.

[0012] Further, in the above sand mold individual identification system, the first shape model and the second shape model can be configured to be those obtained by photographing the pattern at a predetermined location on the surface of the sand mold.

[0013] According to the above configuration, the range of the shape model to be collated can be narrowed, and the collation speed can be improved.

[0014] Further, in the above sand mold individual identification system, a plurality of the first shape models and the second shape models are extracted from one sand mold, and the collation unit first specifies the first shape model that matches using only one of the second shape models, and if the first shape model that matches cannot be specified from the shape model, the first shape model that matches is specified using another second shape model.

[0015] According to the above configuration, it is possible to improve the collation accuracy while suppressing a decrease in the efficiency of the collation operation.

[0016] Further, in the above sand mold individual identification system, a plurality of collation regions having different areas are set from one piece of imaging data for the first shape model and the second shape model, and the collation unit first uses only the collation region having the largest area in the second shape model to identify the first shape model that matches. When the identification of the matching shape model cannot be achieved from the collation region, the first shape model that matches can be identified using other collation regions in the second shape model.

[0017] According to the above configuration, it is possible to improve the collation accuracy while suppressing a decrease in the efficiency of the collation operation.

[0018] Further, in order to solve the above problems, a sand mold data input system according to a second aspect of the present invention includes a first imaging unit that images a pattern on the surface of a sand mold, and a first database that stores, for each sand mold imaged by the first imaging unit, a first shape model obtained from the imaging data in association with the manufacturing conditions of the sand mold.

[0019] Further, in order to solve the above problems, a sand mold data collation system according to a third aspect of the present invention includes a first database that stores, for each sand mold imaged by a first imaging unit, a first shape model obtained from the imaging data in association with the manufacturing conditions of the sand mold, a second imaging unit that images a pattern on the surface of the sand mold at the usage stage during casting, a collation unit that identifies the first shape model that matches the second shape model obtained from the imaging data of the second imaging unit, and a second database that stores, in association with the identification ID assigned to the casting product manufactured using the sand mold, the manufacturing conditions recorded in the first database corresponding to the first shape model identified by the collation unit.

Effects of the Invention

[0020] The sand mold individual identification system of the present invention utilizes the random patterns (shape models of sand grains) that naturally occur on the surface of the sand mold without engraving management codes or the like on the sand mold. As a result, in the sand mold manufacturing line, engraving devices such as laser irradiation devices are no longer required, and it is possible to identify individual sand molds while suppressing equipment costs. In addition, the sand mold data input system and the sand mold data verification system of the present invention can be combined to form the sand mold individual identification system of the present invention.

Brief Description of the Drawings

[0021]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Embodiments for Carrying Out the Invention

[0022] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 is a block diagram schematically showing the flow of the casting line. Here, a casting line for manufacturing a casting product having a cavity inside, that is, a casting line using a master mold and a core as sand molds, is exemplified.

[0023] In the casting line shown in Fig. 1, first, the main mold and the core are manufactured (main mold molding and core molding). During casting, the core is placed in the main mold (core placement), and the paired mold is combined and held in the sand mold with the core placed (frame alignment). Molten cast iron is poured into the frame-aligned sand mold (casting), and after cooling, the sand mold is broken and the cast product is taken out (mold removal). The taken-out cast product is shipped after being inspected after product finishing. Also, the sand mold broken at the mold removal stage can be reused for main mold molding through the processes of raw sand mixing and sand property measurement.

[0024] Conventionally, for the main mold, an identification ID (model, manufacturing date, serial number, etc.) for the final product has been attached (stamped), and it has been possible to identify the individual sand mold by this identification ID. This is because the identification ID attached to the main mold is transferred to the outer surface of the cast product, which is the final product, and is easy to use for the management of the final product. On the other hand, it has not been common to perform individual identification for the core. This is because even if an identification ID is stamped on the core, this identification ID is transferred to the inside (inner surface of the cavity part) of the cast product, which is the final product, and it is difficult to use for the management of the final product. The present invention provides an individual identification method that enables individual identification while suppressing equipment costs for sand molds that have not been individually identified until now.

[0025] The sand mold used for casting is formed by filling sand in a mold and firing it, etc. Therefore, the shapes of sand molds formed using the same mold are almost the same. However, the sand (sand grains) that makes up the sand mold itself is not the same. Each sand grain has different shapes and colors, etc., and different sand grains are mixed in various ratios and distributions to form the sand mold. Therefore, even for sand molds with the same shape, when their surfaces are magnified, they are not composed of the same sand grains. In other words, the surface of the sand mold has a random pattern formed by a collection of many sand grains, and the pattern is different for each sand mold, and there is no completely identical one. Therefore, the shape pattern of the pattern on the sand mold surface can be used as identification information for the sand mold.

[0026] Regarding cores, in addition to the type that is baked (shell cores), there are also those that use a chemical reaction such as an adhesive, called cold boxes or self-hardening molds, and the sand molds of cores are not limited to those formed by baking. In the present embodiment, the case where the core is a shell core is exemplified.

[0027] Hereinafter, a sand mold individual identification system (hereinafter, this system) that performs individual identification based on the shape pattern of the sand mold surface will be described. FIG. 2 is an explanatory diagram for explaining the schematic configuration of this system. Here, the case where the individual identification of the core is performed by this system is exemplified.

[0028] As shown in FIG. 2, this system is roughly composed of a sand mold data input system 10 and a sand mold data verification system 20. The sand mold data input system 10 is a system that photographs a core on a core molding line, associates the shape model (shape pattern) of the photographed core surface with the manufacturing data of the core, and inputs them into a database. The timing for photographing the core in the sand mold data input system 10 may be at least the timing when the manufacturing data for each core is known, and the timing immediately after manufacturing is preferable. The sand mold data verification system 20 is a system that photographs the core to be used on a casting line and verifies the shape model of the photographed core surface with the shape model stored in the database.

[0029] The sand mold data input system 10 includes a manufacturing data acquisition unit 11, a core manufacturing database 12, a surface imaging unit (first imaging unit) 13, a shape model extraction unit 14, and a shape model database (first database) 15. The manufacturing data acquisition unit 11 acquires the manufacturing data of cores (such as manufacturing conditions like temperature, pressure, and firing time) in a core firing machine (an example of a core manufacturing machine), and stores the acquired manufacturing data in the core manufacturing database 12 in association with the identification ID (core ID) of each core. The surface imaging unit 13 images the surface of a predetermined location on the core immediately after manufacturing. The shape model extraction unit 14 performs image processing on the imaging data of the core surface by the surface imaging unit 13, and extracts the shape model of the core surface. The extracted shape model is stored in the shape model database 15 in association with the stored data (core ID and manufacturing data) in the core manufacturing database 12. That is, the stored data of each core in the core manufacturing database 12 is associated with the core ID, manufacturing data, and shape model. In the present embodiment, the shape model obtained from the imaging data of the surface imaging unit 13 and stored in the shape model database 15 corresponds to the first shape model described in the claims.

[0030] The sand mold data verification system 20 includes a shape model database 15, a manufacturing data acquisition unit 21, a master mold manufacturing database 22, a surface imaging unit (second imaging unit) 23, a shape model extraction unit 24, a shape model verification unit (verification unit) 25, and a sand mold database (second database) 26. The manufacturing data acquisition unit 21 acquires the manufacturing data (such as manufacturing conditions like temperature, pressure, and firing time) of the master mold manufactured by a master mold shaping machine in the master mold shaping line, and stores the acquired manufacturing data in the master mold manufacturing database 22 in association with an identification ID. An identification ID (such as model type, manufacturing date, serial number, etc.) for the final product is engraved on the master mold, and this identification ID is also used as the identification ID of each master mold in the master mold manufacturing database 22.

[0031] The surface imaging unit 23 images the surface of a predetermined location on the core (i.e., the core combined with the master mold by sand mold assembly) used during casting. The shape model extraction unit 24 performs image processing on the imaging data of the core surface by the surface imaging unit 23 and extracts the shape model of the core surface. The shape model matching unit 25 matches the shape model extracted by the shape model extraction unit 24 with the shape models stored in the shape model database 15 and identifies the matching shape model. In the present embodiment, the shape model obtained from the imaging data of the surface imaging unit 23 corresponds to the second shape model described in the claims.

[0032] When the shape model matching unit 25 identifies the matching shape model, the manufacturing data corresponding to the shape model is read from the shape model database 15, and the data of the master mold (identification ID and manufacturing data) combined with the core at this time is read from the master mold manufacturing database 22. These are associated and stored in the sand mold database 26. That is, the stored data of each sand mold in the sand mold database 26 is such that the identification ID, the manufacturing data of the master mold, and the manufacturing data of the core are associated. In the sand mold database 26, since the manufacturing data of the core can be traced by being associated with the identification ID, it is not necessary to store the core ID and the shape model.

[0033] In a product (cast product) to which this system is applied on the production line, if a defective product or the like occurs, it is possible to trace all the sand molds (master mold and core) used for casting from the identification ID attached to the product. That is, it is possible to search for the manufacturing history data including the manufacturing data of the sand mold used for the defective product. Also, when there is a product that has malfunctioned during use, it is possible to quickly identify products that may cause the same malfunction and perform inspections and retrievals.

[0034] As described above, in this system, individual identification can be performed using random patterns (shape models of sand grains) that occur naturally on the surface of the cores, rather than serial numbers or the like. That is, engraving processes such as letters and patterns on the cores are not required, and individual identification of the cores can be achieved using the photographed data of the core surfaces. For this reason, engraving devices such as laser irradiation devices are not required in the core manufacturing line, and equipment costs can be suppressed. Note that a photographing device (surface photographing units 13 and 23) for obtaining the photographed data of the core surfaces is necessary, but such a photographing device is relatively inexpensive and requires almost no working time for photographing. Furthermore, it is also easy to incorporate a photographing device, a personal computer (PC) that functions as the shape model extraction units 14 and 24 or the shape model collation unit 25, etc., with almost no change (modification) to the manufacturing line, and the costs for changing the manufacturing line can also be suppressed. In addition, this system also has the merit that it can be easily applied even to small-sized sand molds for which it is difficult to attach identification tags or engravings (it is difficult to secure the surface area for attaching tags or engravings).

[0035] As a method for incorporating the surface photographing units 13 and 23 in this system into the manufacturing line, for example, the following methods can be mentioned. FIG. 3 is an explanatory diagram showing the schematic configuration of a photographing device 30 incorporated as the surface photographing unit 13 in the core manufacturing line. Of course, the photographing device 30 can also be incorporated as the surface photographing unit 23 in the casting line.

[0036] The imaging device 30 can be installed near the outlet of the neutron firing machine in the neutron production line, and includes an imaging table 31, an imaging camera 32, and illumination 33. The imaging table 31 is a table for placing the neutrons to be imaged. The imaging camera 32 is a means for imaging the neutrons. In the imaging device 30, for space saving, the imaging camera 32 is arranged below the imaging table 31, and the neutrons are imaged through a mirror 34. The imaging data obtained by the imaging camera 32 is input into a personal computer having the functions of a shape model extraction unit 14 and a shape model database 15. The illumination 33 is a means for applying illumination light to the neutrons to be imaged. By providing polarization filters for the imaging camera 32 and the illumination 33, it becomes possible to perform imaging with reflection cut. Also, in order to suppress image blurring due to vibration from other equipment, there are methods such as installing a damper on the frame where the imaging camera 32 is installed, or increasing the illumination intensity to shorten the shutter speed.

[0037] For improving the efficiency of the imaging operation, it is preferable that when neutrons are placed at a predetermined position on the imaging table 31, the imaging device 30 detects this and automatically performs imaging. When the neutrons placed on the imaging table 31 are placed at a predetermined position (the position where the camera is in focus, immediately in front of the mirror 34), the switch 35 is turned on and the imaging starts. At this time, the operator may place the neutrons pressing against the location where the switch 35 is located so that the imaging is surely performed. The switch 35 serving as an imaging trigger may be a heat-resistant switch for high-temperature neutrons immediately after firing, or may be a non-contact type switch such as a photoelectric sensor. Such an imaging device 30 can be arranged in a space-saving manner and can be easily arranged without changing the neutron production line.

[0038] The shape model extraction units 14 and 24 basically have the same configuration. They perform shooting of the same location on the core of the object to be photographed, and obtain a shape model through similar image processing. This shape data is preferably obtained by narrowing down a predetermined detection range from the shooting data and performing image processing on the shooting data within this detection range. That is, the shooting range for the sand mold is set as a wide range that includes the external characteristics of the sand mold, and the detection range that becomes the shape data is narrowed down from the external characteristic locations in the shooting range using image processing and the like. Thereby, the shape models obtained by the shape model extraction units 14 and 24 can be surely the same location with respect to the sand mold (the positional deviation of the shape model can be prevented).

[0039] Also, the shape models obtained by the shape model extraction units 14 and 24 are preferably those in which the edge shape of the sand grain pattern in the shooting data is extracted by image processing (see FIG. 4). The shape model extracted in this way uses the shape of the sand grains, and it becomes easier to recognize even if there are color changes or dirt on the core surface during the process. However, using the edge-extracted shape model is not essential in the present invention, and the shooting data itself may be used as the shape model.

[0040] The shape model matching unit 25 performs the matching of the shape model by image processing software, but the algorithm of the image processing software is not particularly limited. As the image processing software of the shape model matching unit 25, for example, pattern matching, feature quantity matching, etc. can be used. Although there are differences in processing speed and robustness, it can also be implemented with open source code.

[0041] FIG. 5 is a flowchart showing an example of a method for matching shape models in the shape model matching unit 25. The shape model matching unit 25 compares the shape model extracted by the shape model extraction unit 24 with the shape model stored in the shape model database 15, that is, one of the shape models extracted by the shape model extraction unit 14 (S1). The comparison result is quantified as a matching score, and the higher the similarity of the compared shape models, the larger the value of the matching score. When the matching score calculated in the comparison of S1 is equal to or greater than a predetermined threshold (YES in S2), the shape model that was the comparison target is put into the candidate list (listed: S3). Thereafter, if there is an uncompared shape model in the shape model database 15 (YES in S3), the processes of S1 to S3 are repeated, and if not, the process proceeds to S5. In S5, the shape model with the highest matching score among the shape models put into the candidate list is specified as the matching shape model.

[0042] In addition, in the step of S2, the threshold is provided to prevent misjudgment and shorten the calculation processing time (if there is no threshold, the search and matching will continue until the matching score becomes almost 0). Also, the listing of the shape models in the step of S3 is not essential in the present invention, but is performed as a record (log) when an error occurs. That is, if there is a large stain or the like on the surface of the sand mold when the sand mold is re-photographed, the matching score for the shape data that should originally match will decrease significantly, and as a result, the matching score for the shape data that should not match will become the maximum value, and there is a possibility of a misjudgment error. The candidate list in the step of S3 is retained as a log when such an error occurs.

[0043] In addition, when implementing this system, almost 100% matching accuracy is required in the matching of shape models (specifying the matching shape models). In the demonstration experiment by the applicant of this application, 100% matching accuracy was obtained in 340 samples.

[0044] The matching accuracy of the shape model can be easily improved by methods such as increasing the number of shape models or increasing the area (number of pixels) of the shape model. However, in these methods, in exchange for the improvement of the matching accuracy, the load on the shape model matching unit 25 increases, and the efficiency of the matching operation also decreases. Therefore, as a method of improving the matching accuracy while maintaining the efficiency of the matching operation, the following (Method 1) or (Method 2) can be considered.

[0045] (Method 1) Extract a plurality of shape models from one sand mold and assign priorities to these multiple shape models. For example, extract two shape models from one sand mold, with one as the main shape model and the other as the sub shape model. In the shape model matching unit 25, first perform a comparison using only the main shape model, and extract the shape model whose matching score is equal to or higher than a predetermined threshold. If there is only one extracted shape model, it is identified as the matching shape model. On the other hand, if there are two or more extracted shape models, a comparison using the sub shape model is performed again only for the sand molds corresponding to the extracted shape models, and the matching shape model is identified by selecting the shape model whose matching score is equal to or higher than a predetermined threshold.

[0046] (Method 2) Set a plurality of matching regions with different areas for the photographed data of one sand mold and assign priorities to these multiple matching regions. For example, set two matching regions from one photographed data, with the smaller area as the main region and the larger area as the sub region. In the shape model matching unit 25, first perform a comparison using the shape model of the main region, and extract the shape model whose matching score is equal to or higher than a predetermined threshold. If there is only one extracted shape model, it is identified as the matching shape model. On the other hand, if there are two or more extracted shape models, a comparison is performed again using the shape model of the sub region only for the sand molds corresponding to the extracted shape models, and the matching shape model is identified by selecting the shape model whose matching score is equal to or higher than a predetermined threshold.

[0047] In both (Method 1) and (Method 2), screening is performed in the first comparison, and shape models with low matching scores (models with no possibility of matching) are excluded. Then, by performing highly accurate matching in the second and subsequent comparisons, the matching accuracy can be improved. Also, in the second and subsequent comparisons, since the number of shape models to be compared is narrowed down by the first screening, a decrease in the efficiency of the matching operation can be suppressed. Usually, the matching score of a matching shape model is a sufficiently large value compared to the matching scores of non-matching shape models. Therefore, in both (Method 1) and (Method 2), it is considered likely that matching shape models are often identified by only the first comparison.

[0048] Among the stored data of the cores in the shape model database 15, the stored data whose matching is confirmed by the shape model matching unit 25 may be stored in the sand mold database 26 and simultaneously deleted from the shape model database 15. As a result, the number of stored data in the shape model database 15 can be suppressed to a certain level, and this can also improve the matching accuracy and speed.

[0049] Also, the sand mold may be stored for a certain period (for example, about 1 to 2 weeks) between manufacturing and use, and may be damaged before use. The stored data of the sand mold that is not used in this way will not be deleted by matching even if it remains in the shape model database 15. Therefore, the stored data in the shape model database 15 may be deleted after a predetermined period has elapsed since it was stored (after the storage period of the sand mold).

[0050] In this system, the mold data input system 10 and the mold data verification system 20 may exist as separate systems. That is, in the case where a producer (casting product producer) who manufactures casting products using molds orders and purchases molds from another party (mold producer), the mold data input system 10 exists on the mold producer side, and the mold data verification system 20 exists on the casting product producer side. In this case, the shape model database 15 in FIG. 1 exists separately for each of the mold data input system 10 and the mold data verification system 20. The casting product producer receives the data to be stored in the shape model database 15 together with the mold purchased from the mold producer, and inputs this data into the shape model database 15 on the mold data verification system 20 side. Thereby, the shape model databases 15 can be virtually made identical by the separate mold data input system 10 and mold data verification system 20.

[0051] The embodiments disclosed this time are illustrative in all respects and are not a basis for restrictive interpretation. Therefore, the technical scope of the present invention is not interpreted only by the above-described embodiments, but is defined based on the description in the claims. Also, all modifications within the meaning and scope equivalent to the claims are included.

Explanation of Reference Numerals

[0052] 10 Mold data input system 11 Manufacturing data acquisition unit 12 Core manufacturing database 13 Surface imaging unit (first imaging unit) 14 Shape model extraction unit 15 Shape model database (first database) 20 Mold data verification system 21 Manufacturing data acquisition unit 22 Master mold manufacturing database 23 Surface imaging unit (second imaging unit) 24 Shape model extraction unit 25 Shape model verification unit (verification unit) 26 Sand mold database (second database)

Claims

Claim 1: A first imaging unit that captures an image of a random pattern formed by a large number of sand grains on the surface of a sand mold immediately after manufacturing; A first database that stores, in association with the manufacturing conditions of the sand mold, a first shape pattern, which is the random pattern obtained from the imaging data of the first imaging unit, for each sand mold imaged by the first imaging unit; A second imaging unit that captures an image of a random pattern formed by a large number of sand grains on the surface of a sand mold used during casting; A collating unit that identifies the first shape pattern that matches the second shape pattern, which is the random pattern obtained from the imaging data of the second imaging unit; A sand mold individual identification system, comprising a second database that stores, in association with an identification ID assigned to a casting manufactured using the sand mold, the manufacturing conditions recorded in the first database corresponding to the first shape pattern identified by the collating unit.

2. The sand mold individual identification system according to Claim 1, wherein the first shape pattern or the second shape pattern is obtained by extracting, through image processing, the edge shape of the sand grain pattern in the imaging data captured by the first imaging unit or the second imaging unit.

3. The sand mold individual identification system according to Claim 1 or 2, wherein the first shape pattern and the second shape pattern are images of patterns at predetermined locations on the surface of the sand mold.

4. The sand mold individual identification system according to any one of Claims 1 to 3, wherein a plurality of first shape patterns and second shape patterns are extracted from one sand mold, and the collating unit first identifies the first shape pattern that matches using only one of the second shape patterns, and if the first shape pattern that matches cannot be identified from the shape pattern, the collating unit identifies the first shape pattern that matches using another second shape pattern.

5. The sand mold individual identification system according to any one of Claims 1 to 3, wherein a plurality of collation regions having different areas are set from one imaging data for the first shape pattern and the second shape pattern, The matching unit first identifies the first shape pattern that matches using only the matching region with the largest area in the second shape pattern, and when the identification of the matching shape pattern cannot be reached from the matching region, the other matching regions in the second shape pattern are used to identify the first shape pattern that matches. A sand mold individual identification system characterized by this.

6. A first photographing unit that photographs a random pattern formed by a large number of sand grains on the surface of a sand mold immediately after manufacturing, for the sand mold immediately after manufacturing; A sand mold data input system characterized by comprising a first database that stores, in association with the manufacturing conditions of the sand mold, the first shape pattern that is the random pattern obtained from the photographed data for each sand mold photographed by the first photographing unit.

7. For the sand mold immediately after manufacturing, for each sand mold photographed by the first photographing unit, a first shape pattern that is a random pattern formed by a large number of sand grains on the surface of the sand mold obtained from the photographed data of the first photographing unit is stored in association with the manufacturing conditions of the sand mold; a first database; A second photographing unit that photographs a random pattern formed by a large number of sand grains on the surface of a sand mold for the sand mold at the usage stage during casting; A matching unit that identifies the first shape pattern that matches the second shape pattern that is the random pattern obtained from the photographed data of the second photographing unit; A sand mold data matching system characterized by comprising a second database that stores, in association with the identification ID assigned to the casting product manufactured using the sand mold, the manufacturing conditions recorded in the first database corresponding to the first shape pattern identified by the matching unit.

Images