CONTROL UNIT AND METHOD FOR EVALUATING IMAGE DATA IN A HOUSEHOLD APPLIANCE

Description

[0001] The invention relates to a household appliance. In particular, the invention relates to a method and a corresponding control unit for evaluating image data relating to an object in a household appliance.

[0002] A household appliance, in particular an oven, can comprise a camera configured to capture image data relating to an object, in particular a food item, in the interior of the household appliance. The image data can be evaluated, e.g., to identify an object in the interior of the household appliance or to determine the cooking status of a food item in the interior of the household appliance. The quality of the image data typically depends on the respective lighting situation and other parameters, such as the degree of soiling, in the interior of the household appliance.

[0003] EP 2 444 733 A2 describes a method for correcting an image of a food item placed in a cooking appliance. WONWOO LEE ET AL: "Real-time Color Correction for Marker-based Augmented Reality Applications," January 1, 2009, pages 32-35, ISBN 978-0-9806098-0-6, describes a method for color correcting markers for augmented reality applications. CN 107 577 352 A describes a method for evaluating a marking on a household appliance.

[0004] This document deals with the technical task of improving the quality of the evaluation of image data with regard to an object in the interior of a household appliance.

[0005] The problem is solved by the subject matter of the independent patent claims. Advantageous embodiments are defined in particular in the dependent patent claims, described in the following description, or illustrated in the accompanying drawings.

[0006] According to one aspect of the invention, a control unit for a household appliance is described according to claim 1. The household appliance comprises an interior space for accommodating at least one object (in particular for accommodating a food item or an item of laundry or dishes) that is processed or handled in the household appliance. The household appliance can be an oven, a refrigerator or freezer, a dishwasher, a washing machine, a dryer, a kitchen appliance, etc. The interior space can be, for example, a cooking chamber, a dishwashing chamber, a cooling chamber, a laundry room, or a drying room.

[0007] The interior may be at least partially illuminated by a lighting unit (particularly a lamp). The lighting situation in the interior of the household appliance may depend on whether there are objects inside (which, for example, cast a shadow) and / or whether the interior is dirty (and possibly the lighting unit is at least partially covered). Furthermore, the lighting situation may be influenced by light shining into the interior from outside. The lighting situation in the interior of the household appliance may therefore vary.

[0008] At least one reference mark can be arranged in the interior of the household appliance, wherein the reference mark has reference values ​​for one or more different properties. The reference mark can comprise one or more different sub-areas with at least partially different reference values ​​for the one or more properties. Alternatively or additionally, the reference mark can comprise a machine-readable code, in particular a quick response code and / or a logo. The reference mark can be arranged on a wall of the interior or on an object in the interior.

[0009] A reference mark can be designed to (visually) display information relating to the one or more properties of the reference mark, in particular relating to the reference values ​​of the one or more properties. For example, the information relating to the reference values ​​can be encoded into the reference mark (e.g., in the form of a machine-readable code). In particular, the reference mark can have a code that includes information relating to the reference values ​​for the one or more different properties of the reference mark 200). Alternatively or additionally, the reference mark can be designed to refer, via the coding of the reference mark, to a data source in which the reference values ​​of the one or more properties (e.g., specific to a particular type of household appliance) are stored. For example, a reference can be made to a specific app area or to a URL (Uniform Resource Locator).

[0010] According to the invention, at least one property of a plurality of reference marks arranged in the interior relates to a color. The control unit is configured to process the image data in order to at least partially color-normalize the image data. For example, the one or more (optical or optically relevant) properties of a reference mark can include: the color value and / or the color temperature of at least one partial area of ​​the reference mark; the color value and / or the color temperature of a white area of ​​the reference mark; the contrast between two or more partial areas of the reference mark, and / or the brightness of at least one partial area of ​​the reference mark.

[0011] The control unit is configured to capture image data relating to the interior of the household appliance using a camera (e.g. a camera of the household appliance). The camera can be arranged in such a way, and / or the capture range of the camera can be such that the one or more reference marks and / or one or more objects in the interior of the household appliance can be captured by the camera. The image data can comprise a single image and / or a temporal sequence of images. The image data can depict or represent an image of the interior. The image can have N x M pixels, where N and M are typically each 500 or more, or 1000 or more.

[0012] Furthermore, the control unit is configured to identify the reference marks in the image data and, based on the image data, to determine actual values ​​for one or more properties of the individual reference marks. In particular, the image represented by the image data can be evaluated with respect to the interior using one or more image processing methods to determine the actual values ​​for one or more properties of the reference mark.

[0013] The control unit is further configured to process the image data depending on the actual values ​​and depending on the reference values. In particular, the control unit can be configured to compare the actual values ​​with the reference values ​​for the one or more properties of the reference marks. During the processing of the image data, object information relating to an object in the interior of the household appliance can then be determined, in particular based on the comparison of the actual values ​​with the reference values, and / or an enhanced image relating to the interior of the household appliance (compared to the image represented by the image data) can be provided.

[0014] By evaluating the image data with respect to one or more reference marks that have known reference values ​​for one or more properties, a reliable and precise evaluation of the image data with respect to the interior of a household appliance can be performed. In particular, it makes it possible to at least partially compensate for the effects of a specific lighting situation in the interior of the household appliance and / or to at least partially normalize the acquired image data. Furthermore, the optical system in the interior of a household appliance can be calibrated to the ambient conditions. This calibration process can be automated at regular intervals and / or triggered manually upon user request.

[0015] The household appliance can be configured to cook a food item in the interior (e.g., by frying or boiling). The household appliance can, in particular, be an oven. The control unit can then be configured to determine, as object information, information relating to the cooking status of a food item arranged in the interior. For example, the degree of browning of a roast in an oven can be determined. By determining and evaluating actual values ​​for one or more (optical or optically relevant) properties of at least one reference mark, the object information relating to the cooking status of a food item can be determined in a particularly precise and robust manner. When correctly determining the degree of browning of an object in the interior of the household appliance, an accurate reproduction of the object's color in the image data is helpful.However, the color reproduction can depend on the lighting situation, soiling of the camera or a material located between the camera and the object. For example, the control unit for a household appliance described here can be designed as a mobile device (e.g. a smartphone or tablet PC) with a camera. If the mobile device is hung in front of an oven in such a way that the camera is aimed at the object in the oven, the front window, which is usually tinted and distorts the color reproduction, is located between the camera and the object. To ensure that the degree of tanning of the object can still be determined correctly, it is advantageous if the image data is processed depending on the actual values ​​and depending on the reference values ​​in order to calculate out the color distortion and thus at least partially color-normalize the image data.

[0016] The control unit can be configured to determine a transformation based on the actual values ​​and the reference values. Furthermore, the control unit can be configured to transform the image represented by the image data with respect to the interior of the household appliance using the transformation to determine a transformed image. The transformation can be different for each household appliance depending on the installation location and / or other production parameters, such as camera properties. The object information and / or the enhanced image can then be determined in a particularly precise and repeatable manner based on the transformed image.

[0017] The control unit can be configured to determine the transformation in such a way that the value of a distance measure (e.g. the mean square deviation) between the actual values ​​and the reference values ​​for the one or more properties of the reference mark is smaller for the transformed image than for the image represented by the image data. For this purpose, for example, the transformation can be adapted iteratively in order to iteratively reduce the value of the distance measure. The optimization method for determining the transformation can be repeated iteratively until a (local) minimum of the value of the distance measure is reached and / or until another termination criterion is met. By means of a transformation determined in this way, the quality of the provided object information and / or of the enhanced image can be further increased.

[0018] The control unit can be configured to determine the position of the reference mark in the interior of the household appliance. The position of the reference mark can be determined based on the image data. Alternatively or additionally, the type of reference mark can be determined based on the actual values ​​for one or more properties. From the type of reference mark, the (prescribed) position for the determined type of reference mark can then be determined, if necessary using a predefined list.

[0019] The image data can then also be processed depending on the position of the reference mark. In particular, the transformation for transforming the image represented by the image data can be determined depending on the position of the reference mark. By taking into account the position of the reference mark, and thus the position of the lighting situation in the interior of the household appliance indicated by the reference mark, the quality of image data processing can be further improved.

[0020] The control unit can be configured to determine or select a household appliance type from a plurality of household appliance types based on the reference mark identified in the image data, in particular based on the actual values ​​for the one or more properties of the reference mark, and based on assignment information. The (predetermined) assignment information can be configured to assign different combinations of one or more reference marks to different household appliance types from the plurality of household appliance types.

[0021] Example types of household appliances include: one or more different models of an oven; one or more different models of a refrigerator or freezer; one or more different models of a dishwasher; one or more different models of a washing machine; one or more different models of a food processor; and / or one or more different models of a dryer.

[0022] The image data can then also be processed depending on the determined appliance type of the household appliance. In particular, the transformation for transforming the image represented by the image data can be determined depending on the appliance type. The appliance type can, for example, be used to display visually relevant properties of the interior of the household appliance (such as reflection properties and / or the color of the interior walls, the arrangement of a lighting unit for illuminating the interior, etc.). Taking this information into account makes it possible to further improve the quality of image data processing.

[0023] According to the invention, the interior of the domestic appliance has a plurality of different reference marks with respective reference values ​​for the one or more properties. The different reference marks are arranged at different positions in the interior of the domestic appliance. The reference marks of the plurality of reference marks can at least partially have different reference values ​​for the one or more properties. Alternatively or additionally, the reference marks of the plurality of reference marks can be at least partially aligned and / or oriented differently. By using different reference marks in a domestic appliance, the quality of the processing of the image data can be further increased. The attachment of the reference marks can be designed such that the reference marks can be removed by a user of the domestic appliance for specific purposes (e.g. pyrolysis function, cleaning), moved to a different position within the cooking chamber orinterior and / or can be replaced.

[0024] According to the invention, the control unit is configured to determine actual values ​​for the one or more properties of the individual reference marks of the plurality of reference marks based on the image data. Furthermore, the control unit is configured to process the image data depending on the actual values ​​and the reference values ​​of the plurality of reference marks. The transformation for transforming the image represented by the image data is determined depending on the actual values ​​and the reference values ​​of the plurality of reference marks. For this purpose, the control unit is configured to determine a plurality of local transformations for the corresponding plurality of positions based on the actual values ​​and on the reference values ​​for the plurality of reference marks at the corresponding plurality of positions (e.g., such that the value of the distance measure is reduced in each case).Furthermore, the control unit is configured to determine the transformation for the image represented by the image data on the basis of the plurality of local transformations by interpolation between the local transformations for the different positions.

[0025] The differently positioned reference marks inside the appliance can thus be used as different support points for determining a transformation for the image data. By taking differently positioned support points into account, the quality of image data processing can be further improved.

[0026] According to a further aspect of the invention, a household appliance according to claim 9 is described, which comprises an interior with a plurality of different reference marks arranged in the interior. Furthermore, the household appliance comprises at least one camera configured to capture image data relating to the interior. Furthermore, the household appliance comprises the control unit described in this document, which is configured to process the image data to determine object information relating to an object in the interior of the household appliance and / or to provide an enhanced image relating to the interior of the household appliance.

[0027] The household appliance may include a user interface (e.g., with a screen). The control unit may be configured to output the object information and / or the enhanced image via the user interface. This can increase the convenience for a user of the household appliance.

[0028] According to a further aspect, a method for processing image data relating to a household appliance is described according to claim 12. The household appliance has several different reference markers in an interior space. A reference marker can be attached to a wall of the interior space or to an object in the interior space. The reference markers each have reference values ​​for one or more properties.

[0029] It should be noted that any aspects of the control unit, the household appliance, and / or the method described in this document can be combined in a variety of ways. In particular, the features of the patent claims can be combined in a variety of ways.

[0030] The invention will be described in more detail below with reference to exemplary embodiments shown in the accompanying drawings. Figur 1a the interior of an oven as an example of a household appliance; Figur 1b exemplary components of a household appliance; Figur 2 an exemplary reference brand; Figur 3a the interior of an oven with several different reference marks; Figur 3b an example lighting situation in the interior of a furnace; Figur 3c an exemplary transformed image of the interior of the furnace; and Figur 4 a flowchart of an exemplary method for processing image data relating to an object in the interior of a household appliance.

[0031] As stated at the beginning, this document deals with the reliable and precise evaluation of image data captured inside a household appliance. In this context, Fig. 1a The interior 110 of an oven 100 is an example of a household appliance. Other examples of household appliances each with an interior include: a refrigerator, a freezer, a freezer, a washing machine, a dryer, a dishwasher, a food processor, etc.

[0032] A food item 112 (as an example of an object) can be placed on a rack 111 in the interior 110 of the oven 100 and cooked in the interior 110 of the oven 100. The interior 110 can be illuminated by a lighting unit 102 to enable a user of the oven 100 to conveniently monitor the progress of the cooking process of the food item 112.

[0033] Furthermore, the oven 100 may comprise at least one camera 103 configured to capture image data relating to the interior 110, in particular relating to the food 112 in the interior 110. Furthermore, the oven 100 comprises, as shown in Fig. 1b Illustrated is a control or computing unit 101 configured to evaluate the image data captured by the camera 103. For example, the control unit 101 can be configured to automatically determine the cooking status of the food 112 based on the image data. Furthermore, the control unit 101 can be configured to output information relating to the evaluated image data, e.g., the determined cooking status of the food 112, via a user interface 104 of the oven 100.

[0034] For example, to capture an image of a food item 112 inside an oven 100, for example, to identify a food item or to determine food properties (e.g., the degree of cooking, the degree of browning, etc.), a camera 103 can be used in the oven 100. The camera 103 can be arranged at the upper edge of the cooking chamber, at one or more corners of the interior 110, and / or at the door to the interior 110.

[0035] The lighting situation in the interior 110 of a household appliance 100 provided by the one or more lighting units 102 typically has inhomogeneous illumination of different sub-areas of the interior 110. The inhomogeneous illumination can be inherently caused by the arrangement of the one or more lighting units 102. Furthermore, an object 112 arranged in the interior 110 can cause one or more sub-areas of the interior 110 to be obscured, so that the one or more sub-areas may be located in a light shadow of the object 112. Furthermore, particularly in the case of an oven 100, significant contamination of the walls bounding the interior 110 can occur. The contamination can, for example, lead to a lighting unit 102 being at least partially obscured by dirt, so that the illumination of the interior 110 deteriorates.In addition, contamination of the interior 110 can change the reflection properties of the walls of the interior 110. Furthermore, the lens of the camera 103 can be contaminated.

[0036] The image data captured by a camera 103 relating to the interior 110 of a household appliance 100 typically depend on the specific illumination of the interior 110. In particular, relatively poor and / or inhomogeneous illumination can result in color distortions, image noise, etc. Impairment of the image data typically also impairs the quality of the evaluation of the image data and / or the quality of the provided images, which can be shared, for example, via a social network.

[0037] This document describes a method that makes it possible to determine a transformation rule, or in short, a transformation, for the image data captured by a camera 103. The transformation rule can be determined based on correction base values ​​at defined support points on the interior wall of the interior 110 and / or on an object 112 in the interior 110 of a household appliance 100. The transformation rule can be applied to the image data to increase the quality of the image data or the images represented by the image data and / or to increase the quality of the evaluation of the image data.

[0038] Fig. 2 shows an exemplary reference mark 200, which can be arranged, for example, on at least one wall of the interior 110 of a household appliance 100. The reference mark 200 can comprise one or more sub-areas 201, each with one or more reference values ​​for one or more properties. The one or more sub-areas 201 can be arranged in a defined manner relative to one another. Exemplary properties of a sub-area 201 are: the size of the area of ​​the sub-area 201; the color of the sub-area 201; the shape of the sub-area 201; and / or a reflectance of the sub-area 201.

[0039] The reference values ​​of the one or more properties of the one or more subregions 201 of a reference mark 200 can be known to the control unit 101 of a household appliance 100 as reference information. Furthermore, the precise arrangement and / or orientation or alignment and / or position of the reference mark 200 within the interior 110 of the household appliance 100 and / or on an object 112 in the interior 110 of the household appliance 110 can be known to the control unit 101 as reference information.

[0040] Fig. 3a shows the interior 110 of the oven 100 from Fig. 1a with multiple reference marks 200 arranged at different locations or positions within the interior space 110. The reference marks 200 can each be arranged or oriented differently. Furthermore, the reference marks 220 can possibly be partially different.

[0041] The control unit 101 can be configured to cause the camera 103 to capture image data relating to the one or more reference marks 200. Furthermore, the control unit 101 can be configured to determine a transformation based on the image data and the reference information relating to the one or more reference marks 200, which transformation can be used to process image data relating to an object 112 in the interior 110 of the furnace 100. In particular, the control unit 101 can be configured to process the image data captured by the camera 103 relating to an object 112 in the interior 110 of the furnace 100 using the transformation. This can improve the quality of the images of the object 112 represented by the image data and / or the quality of the evaluation of the image data.

[0042] Thus, a household appliance 100 is described with an interior 110, with at least one illumination source 102, with at least one camera 103, and preferably with at least two optical reference marks 200 on the walls of the interior 110 and / or on an object 112 in the interior 110. The reference marks 200 are arranged such that the reference marks 200 are located within the detection range of the camera 103. The camera 103 is connected to a computing and / or control unit 101 in the household appliance 100, which can optionally be connected to a remote computing unit, e.g., to a backend server, via a suitable communication connection.

[0043] The at least two marks 200 can already be applied or attached during the production process of the household appliance 100. Alternatively or additionally, a user can be enabled to subsequently attach the marks 200. A mark 200 can be designed such that the mark 200 is uniquely designed for the specific household appliance 200 with regard to its shape and / or pattern. The unique appearance of the mark 200 can enable a unique assignment of the mark 200 to a specific position in the interior 110 of the household appliance 100. In other words, the reference values ​​of the one or more properties of a mark 200 can be configured to enable the control unit 101 to uniquely determine the position of the mark 200 in the interior 110 of the household appliance 100.

[0044] A mark 200 can be designed such that the mark 200 contains one or more features that represent locally defined values ​​of color, contrast, etc. For example, a mark 200 can be designed such that, from the known color application of the mark 200 in one or more sub-regions 201, a distortion in one or more different frequency ranges of the light at the position of the mark 200 can be determined. For example, the mark 200 can have different sub-regions 201 for different fundamental frequencies of the camera 103 (e.g., red, green, blue) in order to determine a transformation for the different fundamental frequencies.

[0045] Fig. 4 shows an exemplary method 400 for processing image data relating to the interior 110 of a household appliance 100. To determine a transformation, image data relating to the interior 110 of the household appliance 100 can be determined (step 401). The image data depends on the individual situation in the interior 110, such as, for example, soiling, the use of a user-specific light source for the lighting unit 102, the current incidence of light from outside into the interior 110, one or more reflections in the interior 110, a shadow cast by one or more objects 112, etc.

[0046] Furthermore, the one or more reference marks 200 can be identified in the image data (step 402). Furthermore, one or more actual values ​​for one or more properties (such as shape, orientation, content, etc.) can be determined for each reference mark 200. The actual values ​​of the one or more properties of a reference mark 200 can be compared with the reference information of the reference mark 200 in the interior 110, and the position of the reference mark 200 in the interior 110 can thus be determined (step 402). Furthermore, based on a comparison between actual properties and reference properties, information relating to the local image properties at the position of the reference mark 200 can be determined (step 403), such as local color temperature by measuring white areas of the reference mark 200; focus adjustment of the camera 102 by measuring the edge transitions of the reference mark 200; local brightness; and / or one or more local measured values ​​for reference colors.

[0047] Furthermore, based on an evaluation of an optical reference mark 200 (such as shape and / or position) and other features of the interior 110 of the household appliance 100, a household appliance type of the household appliance 100 can be determined (step 403). For example, a specific model of a household appliance 100 can be determined and / or it can be determined whether it is an oven, a washing machine, a dishwasher, a refrigerator, etc.

[0048] By fusing the relevant information derived from the household appliance type (e.g., location and type of lighting, interior color, etc.) and / or by determining optical characteristics at the one or more reference marks 200, a suitable transformation rule for an image captured with the camera 103 can be determined (step 404). The transformation rule can be determined in such a way that the transformation rule parameterizes an overall image optimization (e.g., by means of a gamma correction or similar). Alternatively or additionally, one or more transformation rules and / or transformation paths can be determined based on the positions of the one or more reference marks 200.

[0049] The determined transformation (or an inverse of the determined transformation) can be applied to the captured image with respect to the interior 110 (step 405). Furthermore, one or more other known image processing methods can be applied to further enhance the properties of the processed image.

[0050] The support-point-based determination of a transformation rule, with a support point provided by a reference mark 200, enables local optimization of an image captured by a camera 103. Furthermore, it can be ensured that image properties of an image with respect to a cooking item 112 are traced back to defined references. This can prevent, for example, a browning detection from being distorted by an incorrect white balance or the like.

[0051] Fig. 3b shows an exemplary image of the interior 110 of a furnace 100, as it results directly from the image data captured by the camera 103. Based on the image data for the reference marks 200 in the interior 110 of the furnace 100, a transformation can be determined with which the Fig. 3b The resulting transformed image is shown as an example in Fig. 3c shown. From Fig. 3c It can be seen that the transformation can increase the uniformity and / or homogeneity of the illumination within the transformed image. The transformed image can be output, for example, via a user interface 104 of the oven 100 to enable a user of the oven 100 to monitor the condition of the food 112 being cooked in the oven 100. Alternatively or additionally, the cooking condition of the food 112 can be automatically determined with increased accuracy based on the transformed image.

[0052] Fig. 4 1 thus shows a flowchart of an exemplary method 400 for processing image data relating to a household appliance 100, in particular relating to the interior 110 of the household appliance 100. The household appliance 100 has at least one reference mark 200, preferably at least two reference marks 200, in the interior 110. A reference mark 200 has reference values ​​for one or more properties. The method 400 can be executed by a control unit 101 of the household appliance 100.

[0053] The method 400 comprises capturing 401 image data relating to the interior 110 of the household appliance 100 by means of a camera 103 of the household appliance 100. The one or more reference marks 200 and / or the one or more objects 112 that are processed in the interior 110 of the household appliance 100 are typically arranged in the detection range of the camera 103.

[0054] The method 400 further includes identifying 402 the reference mark 200 in the image data. For this purpose, one or more image processing methods, in particular object recognition methods, can be used.

[0055] Furthermore, the method 400 includes determining 403 actual values ​​for the one or more properties of the reference mark 200 based on the image data. For this purpose, one or more image processing methods can be applied, e.g., to determine the color value and / or brightness of a partial area 201 of the mark 200.

[0056] Furthermore, the method 400 comprises processing 404, 405 the image data as a function of the actual values ​​and as a function of the reference values. The image data can be processed, in particular, as a function of a difference or a deviation between the actual values ​​and the reference values. The image data can be processed to determine object information (e.g., the cooking state) with respect to an object 112 (e.g., a food item) in the interior 110 of the household appliance 100 and / or to provide an enhanced image (in particular, an image with improved illumination) with respect to the interior 110 of the household appliance 100.

[0057] The measures described in this document can ensure that the image quality provided by a camera 103 of a household appliance 100 is independent of contamination on the lighting, etc., in the interior 110 of the household appliance 100. Furthermore, variations in the lighting, e.g., caused by obscuring the lighting by a food item 112, external light exposure, or reflections on metallic surfaces of cooking utensils and cookware, can be corrected. Furthermore, a clear standardization of color values ​​in a specific household appliance 100 in the respective current environment of the household appliance 100, e.g., in the user's kitchen, is enabled.

[0058] The present invention is not limited to the exemplary embodiments shown. In particular, it should be noted that the description and the figures are intended only to illustrate the principle of the described control unit, the described household appliance, and / or the described method.

Claims

1. Control unit (101) for a household appliance (100) which has a plurality of different reference marks (200) in an interior (110); wherein the reference marks (200) in each case have reference values of one or more different properties; wherein at least one property of the reference marks (200) relates to a colour; wherein the different reference marks (200) are arranged at different positions in the interior (110) of the household appliance (100); wherein the control unit (101) is designed - to record image data relative to the interior (110) of the household appliance (100) by means of a camera (103); - to identify the plurality of different reference marks (200) in the image data; - to determine actual values of the one or more properties of the individual reference marks (200) of the plurality of reference marks (200) on the basis of the image data; - to determine a plurality of local transformations for the corresponding plurality of positions on the basis of the actual values and on the basis of the reference values of the plurality of reference marks (200) at the corresponding plurality of positions; - to determine a transformation on the basis of the plurality of local transformations, by interpolation between the local transformations for the different positions; - to transform an image represented by the image data relative to the interior (110) of the household appliance (100) by means of the transformation in order to determine a transformed image, wherein the image data is processed in order to normalize the image data at least partially in terms of colour; and - to determine, on the basis of the transformed image, object information relative to an object (112) in the interior (110) of the household appliance (100) and / or an evaluated image relative to the interior (110) of the household appliance (100).

2. Control unit (101) according to claim 1, wherein the control unit (101) is designed - to compare the actual values with the reference values of the one or more properties of the reference marks (200) and - to process the image data as a function of the comparison.

3. Control unit (101) according to one of the preceding claims, wherein the control unit (101) is designed to determine the transformation such that a value of an interval between the actual values and the reference values of the one or more properties of at least one reference mark (200) is less for the transformed image than for the image represented by the image data.

4. Control unit (101) according to one of the preceding claims, wherein the control unit (101) is designed - to determine the positions of the plurality of reference marks (200) in the interior (110) of the household appliance (100); and - to determine the transformation for the transformation of the image represented by the image data as a function of the positions of the plurality of reference marks (200).

5. Control unit (101) according to one of the preceding claims, wherein the control unit (101) is designed - to determine a household appliance type of the household appliance (100) from a plurality of household appliance types on the basis of at least one reference mark (200) identified in the image data, in particular on the basis of the actual values of the one or more properties of at least one reference mark (200) and on the basis of assignment information; wherein the assignment information is configured to assign different combinations of one or more reference marks (200) to different household appliance types from the plurality of household appliance types; and - to process the image data, also as a function of the household appliance type, in particular to determine the transformation for the transformation of the image represented by the image data as a function of the household appliance type.

6. Control unit (101) according to claim 5, wherein the plurality of household appliance types comprises - one or more models of an oven; - one or more models of a refrigerator or freezer; - one or more models of a dishwasher; - one or more models of a washing machine; - one or more models of a kitchen appliance; and / or - one or more models of a dryer.

7. Control unit (101) according to one of the preceding claims, wherein the one or more properties of a reference mark (200) comprise - a colour value and / or a colour temperature of at least one partial region (201) of the reference mark (200); - a colour value and / or a colour temperature of a white area of the reference mark (200); - a contrast between two or more partial regions (201) of the reference mark (200); - a brightness of at least one partial region (201) of the reference mark (200).

8. Control unit (101) according to one of the preceding claims, wherein - the reference marks (200) of the plurality of reference marks (200) have at least partially different reference values of the one or more properties; and / or - the reference marks (200) of the plurality of reference marks (200) are at least partially differently aligned and / or oriented.

9. Household appliance (100) which comprises - an interior (110) for receiving at least one object (112) which is treated or processed in the household appliance (100); - a plurality of different reference marks (200) arranged in the interior (110). - at least one camera (112) which is designed to record image data relative to the interior (110); and - a control unit (101) according to one of the preceding claims, which is configured to process the image data in order to determine object information relative to an object (112) in the interior (110) of the household appliance (100) and / or in order to provide an evaluated image relative to the interior (110) of the household appliance (100).

10. Household appliance (100) according to claim 9, wherein - the household appliance (100) is designed to cook a food in the interior (110); and - the control unit (101) is designed to determine information relative to a cooking state of a food arranged in the interior (110) as object information.

11. Household appliance (100) according to one of claims 9 to 10, wherein - a reference mark (200) comprises a machine-readable code, in particular a quick response code; and / or - a reference mark (200) comprises a logo; and / or - a reference mark (200) has a code which comprises information relative to the reference values of the one or more different properties of the reference mark (200); and / or - a reference mark (200) comprises one or more different partial regions (201) with different reference values of the one or more properties.

12. Method (400) for processing image data relative to a household appliance (100), wherein the household appliance (100) has a plurality of different reference marks (200) in an interior (110); wherein the reference marks (200) in each case have reference values of one or more properties; wherein at least one property of the reference marks (200) relates to a colour; wherein the different reference marks (200) are arranged at different positions in the interior (110) of the household appliance (100); wherein the method (400) comprises - recording (401) image data relative to the interior (110) of the household appliance (100) by means of a camera (103); - identifying (402) the plurality of different reference marks (200) in the image data; - determining (403) actual values of the one or more properties of the individual reference marks (200) of the plurality of reference marks (200) on the basis of the image data; - determining a plurality of local transformations for the corresponding plurality of positions on the basis of the actual values and on the basis of the reference values of the plurality of reference marks (200) at the corresponding plurality of positions; - determining a transformation on the basis of the plurality of local transformations, by interpolation between the local transformations for the different positions; - transforming an image represented by the image data relative to the interior (110) of the household appliance (100) by means of the transformation in order to determine a transformed image, wherein the image data is processed in order to normalize the image data at least partially in terms of colour; and - determining, on the basis of the transformed image, object information relative to an object (112) in the interior (110) of the household appliance (100) and / or an evaluated image relative to the interior (110) of the household appliance (100).