Measuring apparatus and method for measuring a machining tool

Abstract

The invention relates to a measuring apparatus (1) for measuring a machining tool (30), the measuring apparatus (1) comprising: a holding device (2) for holding the machining tool (30), a measuring camera (3) which is or can be oriented towards a tool portion (32), an identification system (4) having an identification camera (5) which is different from the measuring camera (3) and which is or can be oriented towards the held machining tool (30), and a data processing device (6) which is designed to carry out a machine learning algorithm (A), wherein: the machine learning algorithm (A) is trained to identify, on the basis of at least one image (B) captured by the identification camera (5), a tool class (K) to which the imaged machining tool (30) belongs; the holding device (2) and the measuring camera (3) are controlledly adjustable relative to each other so that a number of different measurement processes can be carried out; the data processing device (6) is configured both to automatically select, from the number of measurement processes, a measurement process predefined for the tool class (K) identified by the machine learning algorithm (A) and to control adjustment of the holding device (2) and the measuring camera (3) relative to each other for the carrying out of the selected measurement process.

Description

[0001] P 62206 WO 18 August 2025

[0002] - 1 - TK / TK

[0003] Measuring device and method for measuring a cutting tool

[0004] SCOPE OF APPLICATION AND STATE OF THE ART

[0005] The invention relates to a measuring device for measuring a cutting tool which has at least one cutting edge. The invention also relates to a method for measuring a cutting tool having at least one cutting edge using a measuring device.

[0006] When using modern machine tools and machining centers with automated production processes, it is typically important that the machining tools intended for use, which are primarily cutting tools, are set or measured with high precision. Devices designed for this purpose are often referred to as tool presetting devices or tool setting devices (or simply presetting devices).

[0007] Conventional tool presetting devices have a base to which a tool holder, usually rotatable, is attached for holding the cutting tool to be measured. The tool holder can be a clamping device for holding the cutting tool. Such tool presetting devices often feature a coordinate slide that carries the actual measuring system of the device, allowing the measuring system to be moved multidimensionally using the coordinate slide. High-precision measurements of cutting tools are possible with optical measuring systems. Such an optical measuring system usually comprises an image processing system with a computer and a measuring camera connected to the computer via data transmission, which can be aligned with the cutting tool to be measured.Conventional tool setting devices can have a display and control unit for operation by a user, as well as a computer-aided control unit for controlling a user-selected and user-initiated measurement sequence.

[0008] Typically, to measure a cutting tool, the cutting tool is first inserted into the tool holder, specifically held by the clamping device. Then, using the measuring camera, a tool-fixed zero point can be aligned with an origin of a measuring coordinate system of the tool presetting device. The P 62206 - 2 -

[0009] The measuring camera is typically aligned with the cutting tool in such a way that one of its fields of view captures a section of the cutting tool to be measured, for example, the tip of a drill-shaped cutting tool. A measurement sequence, which the user has previously selected and / or defined as suitable for measuring the cutting tool, is then typically entered into the computer via a keyboard and a screen interface on the display and control unit. For example, if the operator has identified the cutting tool to be measured as a drill, the operator-selected and / or defined measurement sequence can determine the position of the theoretical tip of the drill from the tool section visible in the measuring camera's field of view.This process typically requires manually entering several sequential steps to create a measurement sequence encompassing these steps, similar to a measurement program. The measuring camera is usually moved to different positions and aimed at various sections of the cutting tool during both the definition and execution of the measurement sequence, thus enabling measurements at different points. Manually entering and / or programming all the necessary steps of a measurement sequence can be very time-consuming and requires extensive specialized knowledge from the operator. Therefore, tool measurements are traditionally performed by specially trained and highly qualified personnel.

[0010] TASK AND SOLUTION

[0011] It is an object of the present invention to provide a measuring device for measuring a cutting tool having at least one cutting edge, and a method for measuring such a cutting tool using a measuring device, both of which have improved properties. In particular, subjective operator influences on the tool measurement should be reduced or even completely eliminated.

[0012] This problem is solved by the subject matter of the independent claims. Preferred embodiments are the subject matter of the dependent claims. The wording of all claims is made explicit by reference to the content of this description. P 62206 - 3 -

[0013] A measuring device according to the invention serves for the, in particular automatic, measurement of a cutting tool which has at least one, in particular geometrically defined, cutting edge. In particular, the measuring device is a tool setting device, more specifically a tool presetting device. The measuring device has a holding device for holding the cutting tool to be measured. The holding device can be designed as a receiving device for at least partially or only partially receiving the cutting tool to be measured. The measuring device has a measuring camera. The measuring camera is aligned or can be aligned with a tool section of the cutting tool held by means of the holding device for a measurement sequence. The measuring device has a recognition system which includes a recognition camera that is different from the measuring camera.The recognition camera is aligned with, or can be aligned to, the cutting tool held by the mounting device. The measuring device also includes a data processing unit. The data processing unit is adapted to execute a machine learning algorithm. In particular, the data processing unit includes means adapted to execute the machine learning algorithm. The machine learning algorithm is trained to identify, based on at least one image (especially a digital one) captured by the recognition camera, a tool class to which the depicted cutting tool belongs. Specifically, the machine learning algorithm is trained to classify the cutting tool depicted in the at least one image captured by the recognition camera, particularly according to tool classes.

[0014] The holding device, in particular including the mounted cutting tool, and the measuring camera are adjustable relative to each other in order to execute and / or create a number of different tool-class-specific measurement sequences. The data processing unit of the measuring device is configured and / or programmed to automatically select, from the number of tool-class-specific measurement sequences, a predefined measurement sequence specifically for the tool class identified by the machine learning algorithm. Furthermore, the data processing unit is configured and / or programmed to control the adjustment of the holding device, particularly with the mounted cutting tool, and the measuring camera relative to each other for the execution of the selected measurement sequence.In other words, the data processing unit can be configured and / or programmed to control adjustments to the mounting device and the surveying camera in order to execute the measurement sequence selected based on the tool class identified by the machine learning algorithm. Each selected measurement sequence can have a measurement sequence-specific P 62206 - 4 -.

[0015] The sequence of movements of the mounting device and the surveying camera relative to each other must be predefined.

[0016] In this way, automatic, and in particular fully automatic, measurement of the cutting tool can be enabled, especially by largely or even completely avoiding decisions to be made by the operator, particularly subjective ones. Specifically, the measurement of the cutting tool is made possible without the operator having to classify the cutting tool to be measured according to tool type and / or without manual selection and / or without manual adjustment and / or without manually creating the measurement sequence for each cutting tool. The measurement of the cutting tool can thus be carried out, in particular at least substantially, independently of the operator, which can improve the reproducibility and / or reliability of the measurement. In some cases, particularly fast measurement may be possible.

[0017] In this context, "tool class-specific" can mean "characteristic of, and in particular precisely about, a respective tool class." Alternatively or additionally, "measurement sequence-specific" can mean "characteristic of, and in particular precisely about, a respective measurement sequence."

[0018] The surveying camera may expediently be equipped with an optical measuring system. This optical measuring system may also include an image processing unit for computer-aided evaluation of images captured by the surveying camera. The image processing unit may be, in particular, a functional component of the data processing unit.

[0019] In an embodiment of the invention, a tool class that is recognizable by means of the machine learning algorithm, in particular automatically, corresponds to a tool type of the cutting tool depicted, in particular by means of the recognition camera.

[0020] In a further embodiment of the invention, the machine learning algorithm is trained to recognize whether the depicted cutting tool belongs to at least one of the following, in particular different, tool classes: tool class comprising drilling tools, in particular drills; another tool class comprising milling tools, in particular milling cutters; another tool class comprising turning tools, in particular turning tools; another tool class comprising reaming tools, in particular reamers; another tool class comprising countersinking tools, in particular countersink drills; other P 62206 - 5 -

[0021] The tool class encompasses other tool types, particularly those unknown to the machine learning algorithm. It is conceivable that the machine learning algorithm is also trained to classify cutting tools according to tool classes, particularly automatically, i.e., to create and / or define tool classes.

[0022] In a further embodiment of the invention, the machine learning algorithm is based on deep learning. Alternatively or additionally, the machine learning algorithm can be self-learning capable and / or self-learning.

[0023] In a further embodiment of the invention, the measuring device comprises a drive unit, in particular a first drive unit, which can be controlled by the data processing device. The drive unit, in particular a first drive unit, is designed to adjust the measuring camera and the mounting device relative to each other along an adjustment direction, in particular a first drive direction, of the measuring device. The adjustment direction, in particular a first drive direction, can be vertically oriented, in particular parallel to a direction of gravity.

[0024] In a further embodiment of the invention, the measuring device comprises a drive unit, in particular a second drive unit, which can be controlled by the data processing unit. This drive unit, in particular a second drive unit, is designed to adjust the measuring camera and the mounting device relative to each other along a direction of adjustment of the measuring device, in particular a second adjustment direction. This direction of adjustment can be horizontal, in particular perpendicular to the direction of gravity. In particular, the first and second directions of adjustment are linearly independent of each other. The second direction of adjustment can alternatively or additionally be perpendicular to a rotation axis of the mounting device, in particular a vertical axis.

[0025] In a further embodiment of the invention, the measuring device comprises a drive unit, in particular a third drive unit, which can be controlled by means of the data processing device. This drive unit, in particular a third drive unit, is designed to rotate the mounting device relative to the surveying camera about the mounting device's axis of rotation, in particular a vertical axis.

[0026] In a further embodiment of the invention, the measuring device comprises a swivel drive unit that can be controlled by the data processing unit. The swivel drive unit is designed to swivel the recognition camera about a swivel axis of the measuring device, in particular a horizontal one. The swivel axis can be perpendicular to the first and / or the second adjustment direction of the measuring device. The extension direction of the swivel axis, the first adjustment direction, and the second adjustment direction can be linearly independent of each other.

[0027] Advantageously, the measuring device can have a device base, in particular a frame-like structure and / or a base body of the measuring device, which supports the mounting device, the surveying camera, and the recognition camera. The mounting device can be rotatably adjustable only about its axis of rotation by means of the third drive device. The surveying camera can be vertically adjustable relative to the device base and the mounting device by means of the first drive device. Furthermore, the surveying camera can be horizontally adjustable perpendicular to the axis of rotation relative to the device base and relative to the mounting device by means of the second drive device. In particular, the mounting device is not vertically or horizontally adjustable relative to the device base.

[0028] Advantageously, the recognition camera is arranged such that it looks down onto the held cutting tool from above, relative to the direction of gravity, and / or is directed towards it, particularly to capture the image in the manner of a full-body shot. The recognition camera can be pivotably mounted at the highest point of the measuring device relative to the direction of gravity.

[0029] In a further embodiment of the invention, the measuring device includes an input device designed for defining, in particular manually, and—alternatively or additionally—adapting at least one tool-class-specific measurement sequence. In particular, the input device can be used to define and / or adapt at least one tool-class-specific measurement sequence with regard to future tool measurements. During such a future measurement, the measurement sequence defined and / or adapted by means of the input device can thus be available for selection based on the tool class recognized by the machine learning algorithm. In particular, an existing measurement sequence can be manually corrected using the input device.

[0030] A method according to the invention serves to measure a cutting tool having at least one cutting edge by means of a measuring device. In particular, when carrying out the method, the cutting tool having at least one cutting edge is measured by means of the measuring device. The measuring device used for method P 62206-7 can be a measuring device according to the invention, as described above. The method comprises a step a) in which at least one image is captured by means of a recognition camera, in particular the measuring device, wherein the recognition camera is aligned with the cutting tool to be measured or, in particular, is aligned automatically. The method also comprises a step b) in which a tool class is identified to which the cutting tool depicted in step a) belongs.The tool class is identified according to step b) by executing a machine learning algorithm trained to classify the depicted cutting tool using a data processing device, in particular a data processing device of the measuring device. The method further comprises a step c) according to which a measurement sequence is created, in particular automatically, and / or a predefined measurement sequence is selected, in particular automatically, for the tool class identified in step b) from a number of tool class-specific measurement sequences using the data processing device. Furthermore, according to step c), measurement sequence-specific control commands for the created and / or selected measurement sequence are generated, in particular automatically, using the data processing device.The method also includes a step d) according to which the measurement sequence created and / or selected in step c) is executed, in particular automatically, by adjusting the cutting tool, in particular the held one, and a measuring camera, in particular a measuring camera of the measuring device, relative to each other based on the generated measurement sequence-specific control commands.

[0031] The procedure is expediently implemented by computer, particularly at least stepwise. In particular, steps b) and c) may be implemented by computer, especially on and / or in the data processing equipment. The data processing equipment may be adapted and / or programmed for the execution of steps b) and c), particularly automatically.

[0032] Advantageously, the method may include a step (o), in particular an initial one, in which the cutting tool to be measured is held by means of a holding device of the measuring apparatus. P 62206 - 8 -

[0033] BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Further advantages and features of the invention will become apparent from the claims and from the following description of a preferred embodiment of the invention, which is illustrated with reference to the drawings. In this context, identical reference numerals refer to identical, similar, or functionally equivalent components.

[0035] It is understood that the features mentioned above and those to be explained below can be used not only in the combinations specified, but also in other combinations or on their own, without leaving the scope of the present invention.

[0036] Fig. 1 shows a schematic side view of an embodiment of a measuring device according to the invention,

[0037] Fig. 2 shows a schematic top view of the measuring device according to Fig. 1, and

[0038] Fig. 3 shows a flowchart for carrying out an embodiment of a method according to the invention for measuring a cutting tool having at least one cutting edge by means of a measuring device, wherein the measuring device according to Figs. 1 and 2 can be used for carrying out the method.

[0039] DETAILED DESCRIPTION OF THE EXECUTION EXAMPLES

[0040] A method V is used to measure a cutting tool 30, which has at least one cutting edge 31, by means of a measuring device 1. For example, the measuring device 1 is used to measure the cutting tool 30 according to method V. The measuring device 1 can therefore be used for method V.

[0041] The measuring device 1 can be a tool setting device, in particular a tool presetting device. The measuring device 1 is designed for measuring the cutting tool 30, in particular automatically. In this case, the measuring device 1 is designed to carry out method V in order to measure the cutting tool 30 automatically. P 62206 - 9 -

[0042] The measuring device 1 includes a holding device 2. The holding device 2 is designed to hold the cutting tool 30 to be measured. The holding device 2 can be a receiving device designed to at least partially or only partially receive the cutting tool 30. The holding device 2 can be designed to clamp the cutting tool 30. The holding device 2 can be designed in the form of a chuck. For example, the holding device 2 has a rotary axis C. The cutting tool 30 can be held by the holding device 2 such that a tool rotation axis of the cutting tool 30 is coaxial with the rotary axis C.

[0043] For example, the holding device 2 has a receiving opening, in particular a receiving bore, in which a section of the cutting tool 30, designed complementary to the receiving opening, can be received and is shown in Fig. 1. The receiving opening of the holding device 2 can correspond to a receiving opening of a machine tool spindle. For example, the receiving opening of the holding device 2 can be shaped complementarily, in particular fittingly, to the cutting tool 30 to be held.

[0044] The cutting tool 30 to be measured can be a rotary tool. The cutting tool 30 can essentially consist of a tool chuck, in particular in the form of a shrink-fit chuck, and the actual tool element with at least one cutting edge 31. For example, the tool chuck has a conical section that fits precisely into the receiving opening of the holding device 2, for example in the form of an HSK tool holder, a Morse taper or a steep taper, as well as a

[0045] Clamping section with a centric and essentially cylindrical

[0046] Tool holder opening. The tool element is received with its shank end in the expandable tool holder opening of the tool chuck and secured in a predefinable axial position by clamping. With the aid of the measuring device 1, in particular the setting device, an operator can determine tool dimensions important for machining, for example the tool length, by measurement and, if necessary, change them by adjustment. Accordingly, the measuring device 1 enables the setting of the cutting tool 30. In this case, an axial tool length of the cutting tool 30 can be set. The axial tool length can, for example, correspond to the distance of a tool tip of the tool element relative to the shrink-fit chuck of the cutting tool 30. P 62206 - 10 -

[0047] The measuring device 1 includes a measuring camera 3. The measuring camera 3 can be aligned, or is already aligned, with a tool section 32 of the cutting tool 30, which is held by the mounting device 2, for a measuring sequence to measure the cutting tool 30. In the illustrations of Figures 1 and 2, the measuring camera 3 is not yet aligned with the tool section 32. In particular, at least one cutting edge 31 is present in the tool section 32 of the cutting tool 30. The measuring camera 3 can be a CCD camera.

[0048] The surveying camera 3 is, for example, a component of an optical measuring system of the measuring device 1. The measuring system of the measuring device 1 can include an image processing system whose computing unit is connected to an input device 11 of the measuring device 1. For example, the surveying camera 3 is attached to an arm of a substantially U-shaped measuring system carrier 20. A lighting device 21 is located, for example, on the horizontally opposite arm, so that a cutting tool 30 arranged between the surveying camera 3 and the lighting device 21 can be measured using transmitted light. Such measuring systems are known per se and are therefore not described in detail here.

[0049] The measuring device 1 has a detection system 4. The detection system 4 has a detection camera 5, which is different from the surveying camera 3. The detection camera 5 is aligned with, or can be aligned to, the held cutting tool 30, in this case independently of the alignment of the surveying camera 3.

[0050] The measuring device 1 includes a data processing unit 6. The data processing unit 6 can be at least partially integrated into the recognition system 4. The data processing unit 6 is adapted to execute a machine learning algorithm A. The machine learning algorithm A is trained on training data, in particular comprising a large number of training images of cutting tools 30 of different tool classes K. The machine learning algorithm A is trained to recognize, based on at least one image B captured by the recognition camera 5, a tool class K to which the cutting tool 30 depicted in image B belongs. The data processing unit 6 can be integrated into the computer of the image processing system of the optical measuring system or be designed separately from it. The machine learning algorithm A is based, for example, on deep learning.

[0051] The mounting device 2 and the surveying camera 3 are adjustable relative to each other in a controlled manner. The mounting device 2 and the surveying camera 3 are adjustable relative to each other in such a way (P 62206 - 11 -) that a number of different tool-class-specific measurement sequences can be performed, depending on the control settings.

[0052] The data processing unit 6 is configured to select a measurement sequence from the number of tool-class-specific measurement sequences. The data processing unit 6 is designed to automatically select precisely the measurement sequence that is predefined and / or most suitable for the tool class K identified by the machine learning algorithm A. Furthermore, the data processing unit 6 is configured to control the adjustment of the mounting device 2 and the measuring camera 3 relative to each other in such a way that the selected measurement sequence is executed. "Configured" can be understood, for example, as synonymous with "programmed".

[0053] Each tool class K, which is automatically recognizable by the machine learning algorithm A, corresponds, for example, to a tool type T of the cutting tool 30 depicted in image B by the recognition camera 5. Tool type T can be a drilling tool, in particular a drill bit. Another tool type T can be a milling tool, in particular a milling cutter. Another tool type T can be a turning tool, in particular a turning tool. Another tool type T can be a reaming tool, in particular a reamer. Another tool type T can be a countersinking tool, in particular a countersink drill bit. Other tool types T are conceivable.

[0054] For example, the learning algorithm A is trained to recognize whether the cutting tool 30 depicted in image B belongs to a first tool class K, K1, which represents drilling tools. Alternatively or additionally, the machine learning algorithm A can be trained to recognize whether the tool 30 depicted in image B belongs to a second tool class K, K2, which represents milling tools, in particular milling cutters. Alternatively or additionally, the machine learning algorithm A can be trained to recognize whether the cutting tool 30 depicted in image B belongs to a third tool class K, K3, which represents turning tools, in particular turning tools.Alternatively or additionally, the machine learning algorithm A can be trained to recognize whether the cutting tool 30 depicted by image B belongs to a fourth tool class K, K4, which represents reaming tools, in particular reamers. Alternatively or additionally, the machine learning algorithm A can be trained to recognize whether the cutting tool 30 depicted by image B belongs to a fifth tool class K, K5, which represents countersinking tools, in particular countersink drills. P 62206 - 12 -.

[0055] Alternatively or additionally, the machine learning algorithm A is trained to recognize whether the cutting tool 30 depicted by image B corresponds to another tool class K, KN, where this other tool class K, KN represents other tool types T, in particular other tool types T unknown to the machine learning algorithm A. The tool class K, KN can thus function as a catch-all class if the machine learning algorithm A cannot assign the tool 30 depicted by image B to any other class K, K1 to K5.

[0056] As already mentioned, the measuring device 1 has an input device 11. Using the input device 11, a measurement sequence, particularly one specific to a particular tool class, can be defined and / or adapted. For example, an existing measurement sequence can be corrected using the input device 11. A measurement sequence defined, adapted, and / or corrected using the input device 11 can then be made available for automatic selection by the data processing device 6 for future measurements.

[0057] For example, the measuring device 1 has a first drive unit 7, which can be controlled by means of the data processing unit 6. By means of the first drive unit 7, the surveying camera 3 can be adjusted relative to the mounting device 2 along a first adjustment direction Z of the measuring device 1.

[0058] The measuring device 1, for example, has a second drive unit 8, which can be controlled by means of the data processing unit 6. By means of the drive unit 8, the surveying camera 3 can be adjusted relative to the mounting device 2 along a second adjustment direction X of the measuring device 1.

[0059] The measuring device 1, for example, has a third drive unit 9, which can be controlled by means of the data processing unit 6. In this case, the third drive unit 9 allows the mounting device 2 to be rotated relative to the surveying camera 3 about the axis of rotation C of the mounting device 2.

[0060] The first adjustment direction Z, for example, runs vertically, specifically parallel to the direction of gravity. The second adjustment direction X, for example, runs horizontally, specifically perpendicular to the direction of gravity. The first adjustment direction Z runs parallel to the axis of rotation C, for example. The second adjustment direction X runs perpendicular to the axis of rotation C, for example. P 62206 - 13 -

[0061] The first drive unit 7 and the second drive unit 8 can be encompassed by a slide arrangement of the measuring device 1. The measuring system carrier 20 can be moved multi-axis relative to the mounting device 2 by means of this slide arrangement in order to bring a tool section 32 relevant for measurement into a field of view of the measuring camera 3 by moving the measuring camera 3, or to move the measuring camera 3 out of a region of the mounting device 2, for example, when a fully measured cutting tool 30 is to be replaced or a cutting tool 30 to be measured is to be held.

[0062] The slide arrangement includes, for example, a horizontal slide movable along the second adjustment direction X, which is guided by a horizontally oriented linear guide fixed in a fixture base 22 of the measuring device 1. The fixture base 22 can be a torsionally rigid, heavy, vibration-damped base body. The mounting device 2 can be arranged on a top surface of the fixture base 22. The horizontal slide carries, for example, a column-like structure 23 of the measuring device 1, which can be moved towards or away from the mounting device 2 by means of the second drive device 8. The column-like structure 23 contains, for example, a vertical slide, which is guided by a vertical guide running parallel to the first adjustment direction Z and which is fixed in the column-like structure 23 serving as a support structure.The measuring system carrier 20, which supports the surveying camera 3 and the opposing lighting device 21, is attached to the vertical slide. The vertical slide can be adjusted, for example, by means of the first drive unit 7 along the first adjustment direction Z. By means of controlled movements of the horizontal slide and the vertical slide, the movable components of the optical measuring system, in particular at least the surveying camera 3, can be freely moved to predefinable positions within a vertical XZ plane within the travel paths of the horizontal slide and the vertical slide.

[0063] In embodiments not shown, a further axis of movement for the movable components of the optical measuring system may be provided, namely a further axis (Y-axis) extending perpendicular to the XZ plane. This further axis of movement may define a further adjustment direction that is linearly independent of the first adjustment direction Z and the second adjustment direction X.

[0064] For example, the measuring device 1 has a swivel drive unit 10 which can be controlled by means of the data processing unit 6. In this case, the recognition camera 5 is swiveled about a pivot axis S of the P 62206 - 14 - by means of the swivel drive unit 10.

[0065] Measuring device 1 is pivotable. The pivot axis S runs, for example, horizontally, in this case perpendicular to the first adjustment direction Z and to the second adjustment direction X.

[0066] Method V is, in particular, an automatic method. Method V can be computer-implemented. In particular, steps b) and c) are computer-implemented, specifically on the data processing unit 6. When measuring the cutting tool 30 according to method V, the following procedure can be used:

[0067] In step a) of the procedure, at least one image B can be captured using the recognition camera 5, wherein the recognition camera 5 is or is aligned with the cutting tool 30 to be measured.

[0068] In a subsequent step b) of the procedure, a tool class K is identified, in particular automatically, to which the cutting tool 30 depicted in image B belongs. The tool class K is identified by executing the machine learning algorithm A using the data processing unit 6. The machine learning algorithm A is trained to classify the cutting tool 30 depicted in image B, specifically to assign the cutting tool 30 to the corresponding tool class K based on image B. The machine learning algorithm A can be an object recognition algorithm.

[0069] In a subsequent step c) of the procedure, a predefined measurement sequence for the detected tool class K is selected from a number of tool class-specific measurement sequences using the data processing unit 6 and / or a tool class-specific measurement sequence is created for the detected tool class K. Simultaneously or immediately thereafter, measurement sequence-specific control commands CC are generated for the selected and / or created measurement sequence that is suitable for the detected tool class K.

[0070] In a subsequent step d) of procedure V, the selected and / or created measurement sequence is executed by adjusting the cutting tool 30 and the measuring camera 30 relative to each other based on the generated measurement sequence-specific control commands CC. Specifically, the first drive unit 7, the second drive unit 8, and the third drive unit 9 are controlled in a coordinated manner based on the measurement sequence-specific control commands CC in order to adjust the holding device 2, including the held cutting tool 30, and the measuring camera 3 relative to each other during the execution of the selected and / or created measurement sequence. P 62206 - 15 -

[0071] Prior to step a), in an initial step o), the cutting tool 30 to be measured can be held by means of the holding device 30.

[0072] It is advisable to achieve a successive optimization of the measurement by retraining and / or further training of the machine learning algorithm A.

Claims

P 62206 - 16 - Patent claims 1. Measuring device (1) for measuring a cutting tool (30) having at least one cutting edge (31), wherein the measuring device (1) comprises: a holding device (2) for holding the cutting tool (30) to be measured, a measuring camera (3) which is aligned or alignable for a measuring sequence onto a tool section (32) of the cutting tool (30) held by means of the holding device (2), a recognition system (4) with a recognition camera (5) different from the measuring camera (3) which is aligned or alignable onto the cutting tool (30) held by means of the holding device (2), a data processing device (6) which is adapted for executing a machine learning algorithm (A), wherein the machine learning algorithm (A) is trained to recognize a tool class (K) on the basis of at least one image (B) captured by means of the recognition camera (5),to which the illustrated cutting tool (30) belongs, - wherein the mounting device (2) and the surveying camera (3) are adjustable relative to each other in order to perform and / or create a number of different tool class-specific measurement sequences, - wherein the data processing device (6) is configured to automatically select a predefined measurement sequence from the number of tool class-specific measurement sequences for the tool class (K) recognized by means of the machine learning algorithm (A) and to control an adjustment of the mounting device (2) and the measuring camera (3) relative to each other to execute the selected measurement sequence.

2. Measuring device (1) according to claim 1 , characterized in that each tool class (K) identifiable by means of the machine learning algorithm (A) corresponds to a tool type (T) of the depicted cutting tool (30).

3. Measuring device (1) according to claim 1 or 2, characterized in that the machine learning algorithm (A) is trained to recognize whether the depicted cutting tool belongs to at least one of the following, in particular different, tool classes (K, K1 , K2, K3, K4, K5, KN): Tool class (K, K1) comprising drilling tools, in particular drills; Tool class (K, K2) comprising milling tools, in particular milling cutters; P 62206 - 17 - Tool class (K, K3) comprising turning tools, in particular turning tools; tool class (K, K4) comprising reaming tools, in particular reamers; tool class (K, K5) comprising countersinking tools, in particular countersink drills; tool class (K, KN) comprising other tool types, in particular those unknown to the machine learning algorithm.

4. Measuring device (1) according to one of claims 1 to 3, characterized in that the machine learning algorithm (A) is based on deep learning.

5. Measuring device (1) according to one of the preceding claims, characterized in that the measuring device (1) has a drive device (7) controllable by means of the data processing device (6), in particular a first, for adjusting the surveying camera (3) and the mounting device (2) relative to each other along a, in particular a first, more particularly vertical, adjustment direction (Z) of the measuring device (1).

6. Measuring device (1) according to one of the preceding claims, characterized in that the measuring device (1) has a drive device (8) controllable by means of the data processing device (6), in particular a second, for adjusting the surveying camera (3) and the mounting device (2) relative to each other along an adjustment direction (X) of the measuring device (1), in particular a second, more particularly horizontal and / or perpendicular to a rotation axis (C) of the mounting device (2).

7. Measuring device (1) according to one of the preceding claims, characterized in that the measuring device (1) has a drive device (9) controllable by means of the data processing device (6), in particular a third, for rotating the mounting device (2) relative to the surveying camera (3) about a rotation axis (C) of the mounting device (2), in particular a vertical axis.

8. Measuring device (1) according to one of the preceding claims, characterized in that the measuring device (1) includes a swivel drive device (10) controllable by means of the data processing device (6) for swiveling the recognition camera (5) about a, in particular horizontal, more particularly perpendicular to a first and / or second P 62206 - 18 - The measuring device (1) has a pivot axis (S) extending in the direction of adjustment (Z, X) of the measuring device (1).

9. Measuring device (1) according to one of the preceding claims, characterized in that the measuring device (1) has an input device (11) which is designed for defining and / or adapting at least one tool class-specific measuring sequence, in particular for future measurements, especially manually.

10. Method (V) for measuring a cutting tool (30) having at least one cutting edge (31) by means of a measuring device (1), in particular according to one of the preceding claims, wherein the method (V) comprises the steps: a) capturing at least one image (B) by means of a recognition camera (5) which is or is directed towards the cutting tool (30) to be measured; b) recognizing a tool class (K) to which the cutting tool (30) depicted during step a) belongs by executing a machine learning algorithm (A) trained to classify the depicted cutting tool (30) by means of a data processing device (6);c) Creating a measurement sequence and / or selecting a predefined measurement sequence for the tool class (K) identified during step b) from a number of tool class-specific measurement sequences using the data processing unit (6) and generating measurement sequence-specific control commands (CC) for the created and / or selected measurement sequence; d) Executing the measurement sequence created and / or selected during step c) by adjusting the cutting tool (30) and a measuring camera (3) relative to each other based on the generated measurement sequence-specific control commands (CC).

Images