DISHWASHER, ARRANGEMENT WITH A DISHWASHER AND METHOD FOR OPERATING A DISHWASHER

Description

[0001] The present invention relates to a dishwasher, an arrangement comprising a dishwasher and a computing unit, and a method for operating a dishwasher.

[0002] Known dishwashers include sensors that can record various operating parameters during a wash program. A dishwasher's control unit can adjust one or more wash program parameters depending on the value of an operating parameter to improve the wash result. For example, sensors are known that analyze the wash liquor and thus determine the dirt load in the wash liquor. Depending on the detected dirt load, the wash liquor can be fully or partially replaced with fresh water.

[0003] US 2009 / 0071508 A1 describes a dishwasher with a controllable spray arm that can be controlled to dispense rinsing liquid specifically onto areas with increased soiling.

[0004] DE 10 2012 223 243 A1 describes a household appliance configured for data transmission with a mobile device featuring an integrated camera. The mobile device's camera can capture an image of dirt in or on the household appliance and transmit it to a server, which analyzes the image of the dirt and transmits a cleaning program appropriate for the dirt to the mobile device.

[0005] DE 10 2014 208 861 A1 describes a water-conducting household appliance which has a receiving unit for receiving at least one signal from a device coupled to the household appliance, a number of actuators for carrying out a plurality of programs, a selection unit for selecting a program from the plurality of programs depending on the received at least one signal and a control unit which is designed to control the actuators to execute the selected program.

[0006] US 2019 / 0239716 A1 discloses a dishwasher with an image capture system configured to capture an image of a dishware. A processing unit analyzes the captured image and determines the presence and location of soiling on the dishware. Using a trained convolutional neural network, the cleanliness of the dishware can also be quantified and a soil type classification of the soiling on the dishware can be determined.

[0007] Based on the detected soiling, a cleaning system is activated to clean the soiled area on the dishes. DE 10 2016 106430 A1, JP 2007 064982 A, and EP 2 009 166 A1 disclose other conventional dishwashers.

[0008] Against this background, it is an object of the present invention to provide an improved dishwasher.

[0009] According to a first aspect, a dishwasher, in particular a household dishwasher, is proposed, comprising a control device for executing a washing program for washing soiled dishes. The dishwasher comprises a receiving unit for receiving a sensor signal indicative of a type of dirt adhering to the soiled dishes from a plurality of types of dirt, wherein each of the types of dirt is determined by a dirt parameter matrix in which each position is assigned to a dirt parameter. Furthermore, a determination unit is provided, which is configured to determine the type of dirt adhering to the soiled dishes based on the received sensor signal. The control device is configured to execute the washing program based on the determined type of dirt.

[0010] This dishwasher has the advantage that the control device can optimally adapt the washing program to the detected types of dirt, thus optimizing the washing result.

[0011] The control device can be implemented in hardware and / or software. In a hardware implementation, the control device can be embodied, for example, as a computer or a microprocessor. In a software implementation, the control device can be embodied as a computer program product, as a function, as a routine, as part of a program code, or as an executable object.

[0012] The respective unit, for example, the receiving unit and the detection unit, can be implemented in hardware and / or software. In a hardware implementation, the respective unit can be configured, for example, as a computer or a microprocessor. In a software implementation, the respective unit can be configured as a computer program product, as a function, as a routine, as part of a program code, or as an executable object. Furthermore, the respective unit can form a component of another unit or of the control device.

[0013] The receiving unit comprises, for example, a modem and is coupled or temporarily coupled to at least one sensor. The coupled or couplable sensor is configured to output the sensor signal to the receiving unit.

[0014] The sensor signal is indicative of the type of dirt adhering to the soiled dishes, with a distinction being made between a number of different types of dirt. Each type of dirt is defined by a dirt parameter matrix. The dirt parameter matrix can also be said to form a fingerprint for the respective type of dirt. Each position in the dirt parameter matrix, i.e., each matrix entry, is assigned to a dirt parameter. A dirt parameter is understood, in particular, to be an abstract value that can be calculated, for example, from the sensor signal using a predetermined method or algorithm. In this respect, each dirt parameter can be assigned a predetermined method or algorithm. A predetermined method can also include a machine learning method. The arrangement in a matrix can be advantageous for mathematical operations.

[0015] For example, the type of dirt "ketchup" should be able to be recognized in a photo.

[0016] Ketchup is often reddish in color, which is why it's possible to analyze the red color of stains in the photo. This can be done using an appropriate algorithm that outputs the value "1" for a completely red stain and "0" for a stain where the color red is not present. A high value of the dirt parameter defined in this way is therefore an indication that the dirt type "ketchup" is present. To avoid confusion with other types of dirt, such as red wine, additional dirt parameters must be defined, such as the morphology and / or structure of the stain.

[0017] The detection unit is configured to determine the type of dirt adhering to the washware. In particular, the detection unit can be configured to detect at least one type of dirt for each soiling or stain on the washware.

[0018] For example, as described above, the detection unit determines one or more of the dirt parameters based on the sensor signal. With each additional dirt parameter determined, the determination of the dirt type can become more precise. This can be achieved, in particular, through an exclusion process. For example, based on the dirt parameter defined above (red color), ketchup and red wine can be almost certainly excluded if a certain threshold value for this dirt parameter is undershot.

[0019] For a soiling that includes different types of dirt, such as a mixture of ketchup and mayonnaise, the detection unit can, for example, determine a percentage of each of the types of dirt included. Furthermore, for spatially distributed dirt, a different type of dirt can be determined for different areas on the washware.

[0020] Depending on the detected dirt types, the control device can execute an optimal dishwashing program. Various dishwashing program parameters can be adjusted, such as the water volume, the dishwashing solution temperature, the circulation pump speed, the duration of individual dishwashing program sections such as pre-wash, main wash, final rinse, and / or drying, the amount and / or type of added dishwashing agent, the dosing times for the dishwashing agent, or several different dishwashing agents, such as detergents, enzymes, rinse aid, bleach, and the like. This allows for optimal dishwashing results and minimizes the use of chemicals to support the dishwashing process, as well as water and / or energy consumption.

[0021] In embodiments of the dishwasher, it comprises an automatic dosing system configured for automatically dosing at least one cleaning agent. The automatic dosing system is preferably configured for dosing multiple cleaning agents as needed. The control device of the dishwasher is configured, in particular, to control the automatic dosing system.

[0022] According to the invention, the sensor signal comprises image information of the soiled dishes.

[0023] The image information is, for example, a digital photo of the soiled dishes, which was captured by a camera on the dishwasher or by an external camera and transmitted to the receiving unit.

[0024] Based on the image information, a soiling type can be determined. For example, the appearance of soiling on the dishes, i.e., geometric and / or morphological factors of soiling, are taken into account. Furthermore, a soiling type can be indirectly determined based on the shape and / or material of the dishes, for example, if a liquid dish, such as soup, is served in a bowl or similar. The shape and / or material of the dishes can also be determined based on the image information.

[0025] According to the invention, the determination unit comprises an image processing unit which is configured to determine a value for at least one of the dirt parameters contained in the dirt parameter matrix as a function of the image information.

[0026] The image processing unit may in particular comprise a graphics processor and / or a neural network.

[0027] According to the invention, a database unit, a generation unit and a comparison unit are provided, wherein the database unit is configured to store a dirt parameter matrix for each of the plurality of dirt types, wherein a respective dirt parameter matrix for each of the dirt parameters comprises a predetermined value or value range, wherein the generation unit is configured to generate a current dirt parameter matrix depending on the determined value for the at least one dirt parameter, wherein the comparison unit is configured to compare the generated current dirt

[0028] Parameter matrix with at least a subset of the dirt parameter matrices stored in the database unit and for outputting a comparison result, and wherein the determination unit is configured to determine the type of dirt adhering to the soiled items to be washed as a function of the comparison result.

[0029] The dirt parameter matrices stored in the database unit form a reference whose entries, i.e., the dirt parameter values, are typical values ​​for a specific type of dirt. These reference matrices are provided, for example, by the dishwasher manufacturer.

[0030] For example, a dirt parameter matrix will be determined for a piece of washware soiled with exactly one type of dirt. It is advantageous if the dirt type is not mixed with other types of dirt. In order to take into account a certain scatter of individual values ​​for dirt parameters, which may arise, for example, from the lighting situation when capturing the image of the washware, an interval and / or a tolerance range or error range can be specified for a type of dirt in the respective dirt parameter matrix instead of a single value. The interval or tolerance range can be determined empirically in particular by determining a respective value for a dirt parameter for a large number of different sensor signals, each of which was recorded from a piece of washware with exactly one type of dirt, and then carrying out a statistical evaluation of the large number of values.

[0031] In embodiments, it can be provided that different dirt parameter matrices are stored for different combinations of dirt type and material of the washware, since, for example, a shape of a stain can depend on the material of the washware, such as plastic, glass, metal or porcelain.

[0032] The generation unit generates a current dirt parameter matrix, the values ​​of which for the dirt parameters are determined or calculated depending on the received sensor signal.

[0033] The comparison unit compares the generated current dirt parameter matrix with one or more of the stored dirt parameter matrices. In this case, comparison is understood to mean, in particular, that each entry in the current dirt parameter matrix is ​​compared with the corresponding entry of exactly one of the stored dirt parameter matrices that corresponds exactly to one type of dirt. For each of these comparisons of individual dirt parameter values, an individual comparison value is determined that represents a measure of agreement. The measurement system used here can be designed in different ways. In particular, it can be provided that the measurement system is different for different dirt parameters in a dirt parameter matrix. Furthermore, the measurement system can be different for the same dirt parameters for different types of dirt, i.e., for different dirt parameter matrices.In this way, a weighting for particularly characteristic dirt parameters for individual dirt types can be achieved.

[0034] Based on all individual comparison values, the comparison result is output based on the comparison of the current dirt parameter matrix with exactly one of the stored dirt parameter matrices. The comparison result is particularly indicative of whether the respective type of dirt is present on the washware or not. For example, in the case of identical matrices, a 100% match can be output as the comparison result. It should be noted that an individual comparison value is not necessarily required for each dirt parameter in order to determine the comparison result. If only a few individual comparison values ​​are available, this can be clarified, for example, by means of an error or confidence interval based on the comparison result. The error or confidence interval is determined by the comparison unit and output with the comparison result.

[0035] The detection unit determines which type of dirt is present on the washware based on the comparison result. For example, the detection unit records the comparison result of the comparison unit for each comparison performed, preferably for each of the stored dirt parameter matrices. The detection unit is configured, for example, to detect multiple dirt types. Determining whether a particular type of dirt is present on the washware can, in turn, be done using different measurement systems.

[0036] For example, if there is a 100% match, it is determined that the corresponding type of dirt is present on the dishes.

[0037] According to a further embodiment of the dishwasher, a learning unit is provided which is configured to adapt at least one of the dirt parameter matrices stored in the database unit as a function of specific values ​​for at least one dirt parameter.

[0038] One could also say that the learning unit is configured to optimize at least one stored dirt parameter matrix. For example, this can be done within a learning mode in which a user specifies dishes soiled with specific dirt types. Alternatively, during regular dishwasher operation, the user can be asked to confirm a specific dirt type. By regularly querying this type of information, the determination of the dirt type can be continuously improved over time.

[0039] According to a further embodiment of the dishwasher, the sensor signal comprises information relating to a rinsing solution used to rinse the soiled dishes.

[0040] For example, the dishwasher has a wash solution sensor that detects the turbidity of the wash solution during the wash program. The turbidity of the wash solution can be included in the dirt parameter matrix for a specific type of dirt as a dirt parameter. This further improves the determination of the dirt type.

[0041] In addition to the turbidity of the rinsing liquor, for example, a particle size of particles dissolved in the rinsing liquor, a conductivity of the rinsing liquor and / or a pH value of the rinsing liquor can be recorded and output as a sensor signal.

[0042] According to a further embodiment of the dishwasher, the sensor signal comprises information relating to a food prepared and / or consumed by a user of the dishwasher in a specific time interval before the start of the washing program.

[0043] For example, the dishwasher is networked with a food processor, a cooktop, and / or an oven, or with a digital cookbook and / or calendar used by the user. Such networking can be achieved via Wi-Fi, LAN, or Bluetooth®. The specific time interval can be, for example, a few hours or a day.

[0044] According to a further embodiment of the dishwasher, the control device is designed to adapt the execution of a running washing program.

[0045] The adjustment is preferably carried out as a function of a sensor signal received during the implementation, such as a sensor signal received from a turbidity sensor.

[0046] According to a further embodiment of the dishwasher, a camera is arranged on the dishwasher, wherein the camera is configured to detect an optical sensor signal of the items to be washed and to output the detected optical sensor signal to the receiving unit.

[0047] The camera is arranged on the dishwasher in such a way that it detects the optical sensor signal of the items to be washed when the dishwasher is loaded with the items to be washed.

[0048] According to a further embodiment of the dishwasher, a statistics unit is provided which is configured to determine a frequency distribution which comprises a frequency of occurrence of different types of dirt, wherein the determination unit is configured to determine the type of dirt adhering to the soiled items to be washed as a function of the frequency distribution.

[0049] The statistics unit determines the frequency distribution, particularly depending on the types of dirt detected and over longer periods of time. This means that the statistics unit saves each type of dirt detected, preferably together with a respective timestamp of the time of detection. This results in a certain statistical probability of occurrence for each type of dirt. The frequency distribution can be different for different users due to different preferences. The frequency distribution can also be used to determine patterns regarding the temporal occurrence of certain types of dirt. One example of a pattern is that fish and potatoes are eaten on Fridays, which results in certain types of dirt. Another example of a pattern is that wine is consumed primarily in the evening.The frequency distribution can be advantageous when two types of dirt are identified with the same uncertainty during a single investigation, i.e., when their comparison results are the same. In this case, for example, the dirt type that occurs more frequently in the frequency distribution can be selected because its probability of occurrence is higher.

[0050] In embodiments, a correlation unit may be provided, which may in particular also be a component of the statistics unit. The correlation unit is configured to provide a correlation between different types of dirt and / or to determine it based on the frequency distribution determined by the statistics unit.

[0051] For example, a correlation can be determined between ketchup and fat, since fatty foods, such as fries or grilled sausages, are preferably consumed with ketchup.

[0052] According to a further embodiment of the dishwasher, a communication unit is provided which is configured to communicate with an external unit, in particular with an external kitchen appliance, a mobile device, and / or a server.

[0053] The server is, for example, a server of a service provider and / or a manufacturer of the dishwasher. The server can be queried to provide dirt parameter matrices for different dirt types. In embodiments, the communication unit can transmit the received sensor signal to the server, which evaluates it and reports the determined dirt types back to the dishwasher.

[0054] According to a further embodiment of the dishwasher, a user interface is provided for detecting a user input as a function of a determined type of dirt by the detection unit.

[0055] This is advantageous because, for example, the user can be questioned if the recognition is uncertain. The user interface can be implemented, in particular, by an application installed on the user's mobile device.

[0056] The respective unit, for example, the database unit, the comparison unit, the determination unit, the image processing unit, the generation unit, the learning unit, the statistics unit, and / or the correlation unit, can be implemented in hardware and / or software. In a hardware implementation, the respective unit can be configured, for example, as a computer or a microprocessor. In a software implementation, the respective unit can be configured as a computer program product, as a function, as a routine, as part of a program code, or as an executable object.

[0057] According to a second aspect, an arrangement comprising a dishwasher, in particular a household dishwasher, and a computing unit is proposed, wherein the dishwasher and the computing unit are configured for data communication with one another. The dishwasher comprises a control device for executing a washing program for washing soiled dishes. The computing unit comprises a receiving unit for receiving a sensor signal indicative of a type of dirt adhering to the soiled dishes from a plurality of types of dirt, wherein each of the types of dirt is determined by a dirt parameter matrix in which each position is assigned to a dirt parameter. A determination unit is provided which is configured to determine the type of dirt adhering to the soiled dishes depending on the received sensor signal.The control device is designed to carry out the rinsing program depending on the type of dirt detected.

[0058] The processing unit receives the sensor signal, in particular, from the dishwasher. Alternatively, the processing unit can receive the sensor signal from a device external to the dishwasher, for example, a mobile device belonging to the dishwasher user.

[0059] Data communication between the dishwasher and the computing unit takes place, in particular, via the internet, with a communication connection being established, for example, via Wi-Fi, mobile communications, a VPN, and the like. The computing unit preferably comprises a server. The computing unit can also be embodied as an instance of an application on a server.

[0060] Advantageously, the computing unit can have a very high computing power, which is why the type of dirt can be determined in a short time based on the sensor signal using complex algorithms.

[0061] The embodiments and features described for the proposed dishwasher according to the first aspect apply accordingly to the proposed arrangement of dishwasher and computing unit. In particular, the various units, such as the determination unit, the image processing unit, the database unit, the generation unit, the comparison unit, the learning unit, the statistics unit, etc., can be configured as a component of the computing unit.

[0062] The computing unit transmits, for example, the determined type of dirt to the control device, whereupon the control device carries out the rinsing program depending on the determined type of dirt.

[0063] According to a third aspect, a method for operating a dishwasher, in particular a household dishwasher, with a control device for carrying out a washing program for washing soiled dishes is proposed. In a first step, a sensor signal indicative of a type of dirt adhering to the soiled dishes from a plurality of types of dirt is received, wherein each of the types of dirt is determined by a dirt parameter matrix in which each position is assigned to a dirt parameter. In a second step, the type of dirt adhering to the soiled dishes is determined as a function of the sensor signal. In a third step, the washing program is carried out as a function of the determined type of dirt.

[0064] According to the invention, this method is carried out using a dishwasher according to the first aspect. The method can also be carried out correspondingly using the arrangement comprising a dishwasher and a computing unit according to the second aspect, even if such a method is not part of the invention defined in the claims.

[0065] The embodiments and features described for the proposed dishwasher apply accordingly to the proposed method.

[0066] According to a fourth aspect, a computer program product is proposed which comprises instructions which, when the program is executed by a computer, cause the computer to carry out the method according to the third aspect.

[0067] A computer program product, such as a computer program means, can be provided or delivered, for example, as a storage medium, such as a memory card, USB stick, CD-ROM, DVD, or in the form of a downloadable file from a server in a network. This can be done, for example, in a wireless communications network by transmitting a corresponding file with the computer program product or the computer program means.

[0068] Further possible implementations of the invention also include combinations of features or embodiments described above or below with respect to the exemplary embodiments that are not explicitly mentioned. In this case, the person skilled in the art will also add individual aspects as improvements or additions to the respective basic form of the invention.

[0069] Further advantageous embodiments and aspects of the invention are the subject of the dependent claims and the exemplary embodiments of the invention described below. The invention will be explained in more detail below using preferred embodiments with reference to the accompanying figures. Fig. 1 shows a schematic perspective view of an embodiment of a dishwasher; Fig. 2 shows a general example of a dirt parameter matrix; Fig. 3 shows specific examples of three dirt parameter matrices; Fig. 4 shows a schematic block diagram of an embodiment of a dishwasher, Fig. 5 shows a schematic block diagram of an exemplary arrangement with a dishwasher and a computing unit; and Fig. 6 shows a schematic block diagram of an embodiment of a method for operating a dishwasher.

[0070] In the figures, identical or functionally equivalent elements have been given the same reference numerals unless otherwise stated.

[0071] The Fig. 1 shows a schematic perspective view of an embodiment of a dishwasher 1, which is designed here as a household dishwasher. The household dishwasher 1 comprises a washing container 2, which can be closed by a door 3, in particular in a watertight manner. For this purpose, a sealing device can be provided between the door 3 and the washing container 2. The washing container 2 is preferably cuboid-shaped. The washing container 2 can be arranged in a housing of the household dishwasher 1. The washing container 2 and the door 3 can form a washing chamber 4 for washing dishes 30.

[0072] Door 3 is in the Fig. 1shown in its open position. The door 3 can be closed or opened by pivoting about a pivot axis 5 provided at a lower end of the door 3. With the help of the door 3, a loading opening 6 of the washing container 2 can be closed or opened. The washing container 2 has a base 7, a ceiling 8 arranged opposite the base 7, a rear wall 9 arranged opposite the closed door 3 and two side walls 10, 11 arranged opposite one another. The base 7, the ceiling 8, the rear wall 9 and the side walls 10, 11 can be made, for example, from a stainless steel sheet. Alternatively, the base 7 can, for example, be made from a plastic material.

[0073] The household dishwasher 1 further comprises at least one dishware receptacle 12 to 14. Preferably, several, for example three, dishware receptacles 12 to 14 can be provided, wherein the dishware receptacle 12 can be a lower dishware receptacle or a lower basket, the dishware receptacle 13 can be an upper dishware receptacle or an upper basket, and the dishware receptacle 14 can be a cutlery drawer. Fig. 1 As further shown, the dishware receptacles 12 to 14 are arranged one above the other in the washing container 2. Each dishware receptacle 12 to 14 can be selectively moved into or out of the washing container 2. In particular, each dishware receptacle 12 to 14 can be pushed or moved into the washing container 2 in an insertion direction E and pulled or moved out of the washing container 2 in an extension direction A, opposite to the insertion direction E.

[0074] Furthermore, a control device 15 for carrying out a washing program for washing soiled items 30 is arranged on the door 3. In this example, the control device 15 comprises a receiving unit 16 (see Fig. 4 ) for receiving a sensor signal indicative of a dirt type SA, SB, SC of a plurality of dirt types SA, SB, SC adhering to the soiled wash ware 30. Each of the dirt types SA, SB, SC is represented by a dirt parameter matrix MXA, MXB, MXC (see Fig. 3 ). The dirt is, for example, grease SA, rice residues SB and a dark sauce SC. Furthermore, the control device 15 has a detection unit 20 (see Fig. 4) which is configured to determine the type of dirt SA, SB, SC adhering to the soiled washware 30 based on the received sensor signal. The control device 15 executes the wash program based on the determined type of dirt SA, SB, SC in order to achieve an improved cleaning result.

[0075] Fig. 2 shows a general example of a dirt parameter matrix SMX. The illustrated general form of the dirt parameter matrix SMX has n×m entries, where n is the number of rows and m is the number of columns. Each of the entries a_11, a_12, ..., a_1m, a_21,... a_n1,... a_nm represents a dirt parameter.

[0076] For example, dirt parameters a_11 - a_nm can be divided into different categories or classes. An example of such a division is shown in Table 1 below. The examples of dirt parameters a_11 - a_nm shown in Table 1 are based on an optical sensor signal that provides image information of the soiled washware 30 (see Fig. 1 ). The image information is received by the receiving unit 16, in particular as a digital image comprising a plurality of pixels. Those pixels that exhibit contamination are referred to as dirty pixels in Table 1. Table 1: category Example 1 Example 2 Spatial distribution Number of dirt pixels in an edge area of ​​the dishes Average distance of the dirt pixels from the center of the dishes morphology Number of connected areas of dirt pixels Number of pixels of the largest contiguous area Color Average green value of all dirt pixels Standard deviation of the mean green value texture Average gray value of all dirt pixels after the image has been transformed into a grayscale image Density of dirt pixels whose gray value is above the average gray value

[0077] It should be noted that the categories and specific examples in Table 1 are only exemplary and not exhaustive.

[0078] For each dirt parameter a_11 - a_nm, for example, a determination rule can be predetermined that specifies how the respective dirt parameter a_11 - a_nm is determined based on the digital image of the washware 30. Different dirt parameters a_11 - a_nm can be completely independent of one another, such as "average distance of the dirt pixels from a center of the washware" and "average green value of all dirt pixels," or they can exhibit interdependencies, such as "average green value of all dirt pixels" and "standard deviation of the average green value."

[0079] For a specific type of dirt, the various dirt parameters a_11 - a_nm have characteristic values, meaning the values ​​lie within a specific range that can be determined empirically, for example. For example, rice grains have a specific size, such as 1 - 10 mm, and shape, such as elliptical. These values ​​depend, for example, on the viewing angle and the type of rice. Spaghetti has an elongated, connected, curvilinear structure. For each type of dirt, characteristic properties can be used to combine a plurality of dirt parameters a_11 - a_nm into a dirt parameter matrix SMX such that the characteristic properties can be differentiated. This means that the types of dirt can be identified based on the values ​​of the dirt parameters a_11 - a_nm.

[0080] Fig. 3shows specific examples for three dirt parameter matrices MXA, MXB, MXC for three dirt types SA, SB, SC (see Fig. 1 ), whereby the dirt parameter matrices MXA, MXB, and MXC contain only four dirt parameters for clarity. An example of a determined current dirt parameter matrix MXi is also shown.

[0081] For example, soil types SA, SB, SC are dark sauce SA, mayonnaise SB, and chocolate sauce SC. The sensor signal is a digital image of the soiled items 30 (see Fig. 1 ). The dirt parameters a_11 - a_nm (see Fig. 2 ) have, for example, the meaning of a spatial distribution of dirt pixels on the dishes, an average brightness of the dirt pixels, a homogeneity of an area formed by the dirt pixels and a number of dirt pixels whose brightness is below a threshold value normalized to the total number of dirt pixels.

[0082] The dirt parameter matrices MXA, MXB, MXC are, for example, characteristic of the respective dirt type SA, SB, SC. ​​For each of the defined dirt parameters a_11 - a_nm, they contain a value range that was determined, for example, through empirical measurements. In such a measurement, the dirt parameter matrix MXA is determined, for example, based on a sensor signal from a dishware 30 that is only soiled with dirt type SA. These measurements are preferably performed by a manufacturer of the dishwasher 1, and the dirt parameter matrices MXA, MXB, MXC are provided.

[0083] During operation of the dishwasher 1, the sensor signal of the soiled washware 30 is detected. In particular, a photo of the washware 30 is taken and transmitted to the receiving unit 16. Based on the sensor signal, the current dirt parameter matrix MXi is determined by determining the value of each dirt parameter a_11 - a_nm based on the sensor signal.

[0084] By comparing the current dirt parameter matrix MXi with the dirt parameter matrices MXA, MXB, MXC, it is possible to determine in particular which dirt type SA, SB, SC is present on the wash ware 30. The comparison is carried out in particular by a comparison unit 24 (see Fig. 4 ). For example, a probability is determined for each of the soil types SA, SB, and SC.

[0085] For example, comparing the current dirt parameter matrix MXi with the dirt parameter matrix MXA shows that all values ​​of the current dirt parameter matrix MXi lie within the respective value range of the dirt parameter matrix MXA. This results in a very high probability that dirt type SA is present on the washware 30.

[0086] The comparison with the other dirt parameter matrices MXB and MXC shows deviations in at least one dirt parameter a_11 - a_nm. The probability that dirt type SB is present on the washware 30 is very low, since all parameter values ​​lie outside the respective value range of the dirt parameter matrix MXB. The probability that dirt type SC is present on the washware 30 is moderate, since two parameter values ​​lie outside the respective value range of the dirt parameter matrix MXC.

[0087] This comparison result can be interpreted in different ways. A first interpretation could be that only the dirt type SA is present on the washware 30. A second interpretation could be that the dirt type SA is present with a small amount of the dirt type SB on the washware. The interpretation of the comparison result is determined in particular by an identification unit 20 (see Fig. 4 ) is carried out.

[0088] Fig. 4 shows a schematic block diagram of an embodiment of a dishwasher 1. This is, for example, the household dishwasher 1 of Fig. 1The household dishwasher comprises a control device 15, a receiving unit 16, a camera 17, a user interface 18, a detection unit 20, which includes an image processing unit 21, a generation unit 23, and a comparison unit 24, a database unit 22, a learning unit 25, a statistics unit 26, and a communication unit 27.

[0089] The receiving unit 16 is designed to receive a sensor signal from soiled washware 30 (see Fig. 1). The sensor signal can be provided by the camera 17 in the form of an image or photo of the wash ware 30. Alternatively or additionally, the sensor signal can be received by an external unit 50, for example a mobile device of the user. Furthermore, the sensor signal can be received by the communication unit 27 and provided to the receiving unit 16. The receiving unit 16 transmits the received sensor signal to the detection unit 20. The detection unit 20 is configured to detect a type of dirt SA, SB, SC adhering to the wash ware 30 (see Fig. 1 ) to determine.

[0090] In this example, the sensor signal comprises an image of the washware 30. The received sensor signal is processed by the image processing unit 21. In particular, a value is determined for each dirt parameter a_11 - a_nm. The generation unit 23 generates a current dirt parameter matrix MXi from the determined values ​​(see Fig. 3 ). The comparison unit 24 compares the current dirt parameter matrix MXi with stored dirt parameter matrices MXA, MXB, MXC (see Fig. 3), which it retrieves from the database unit 22. Preferably, the current dirt parameter matrix MXi is compared with each stored dirt parameter matrix MXA, MXB, MXC. As a comparison result, for example, a probability is determined that the dirt type SA, SB, SC represented by the respective stored dirt parameter matrix MXA, MXB, MXC is present on the washware 30. Based on the comparison result, the determination unit 20 determines which of the dirt types SA, SB, SC is present on the washware 30 and outputs this to the control device 15. The control device 15 then executes a wash program that is specifically adapted to the existing dirt types SA, SB, SC.

[0091] Using the user interface 18, the user of the household dishwasher 1 can be informed about detected dirt types SA, SB, SC and / or the user can be prompted to confirm a detected dirt type SA, SB, SC. ​​Such feedback from the user can advantageously be used by the learning unit 25 to adapt the dirt parameter matrices SA, SB, SC stored in the database unit 22, so that the determination of the dirt types SA, SB, SC is improved in the future.

[0092] The statistics unit 26 is configured to determine a frequency distribution based on the determined dirt types SA, SB, SC, which contains a statistical probability for the occurrence of certain dirt types SA, SB, SC for future determinations. In particular, a correlation between different dirt types SA, SB, SC, as well as a correlation between a time of occurrence and a dirt type SA, SB, SC, can be determined, which can improve the determination of the dirt types SA, SB, SC in the future.

[0093] The communication unit 27 is configured, in particular, to communicate with other household appliances of the user, such as a food processor, a cooking appliance, and / or a refrigerator, or with applications such as a digital cookbook or calendar of the user. Additional information regarding an expected dirt type SA, SB, SC can be received and evaluated from these external devices in order to improve the determination of the dirt types SA, SB, SC.

[0094] Fig. 5 shows a schematic block diagram of an exemplary arrangement with a dishwasher 1, for example the household dishwasher 1 of Fig. 1, and a computing unit 100. The dishwasher 1 and the computing unit 100 are configured to communicate with each other via a communication connection DATA. The computing unit 100 is designed, for example, as a server and is accessible via the Internet. The dishwasher 1 therefore comprises a network interface, such as a WLAN module, so that it can establish a communication connection with the server 100 via the user's Internet access. With this arrangement, the dishwasher 1 can have a relatively simple structure. In particular, it may be sufficient if the dishwasher has a control device 15 (see Fig. 1 or 4 ). The receiving unit 16 (see Fig. 4 ) and Investigation Unit 20 (see Fig. 4 ) are included in particular by the server 100. Further optional units, such as an image processing unit 21 (see Fig. 4), a generating unit 23 (see Fig. 4 ), a comparison unit 24 (see Fig. 4 ), a database unit 22 (see Fig. 4 ), a learning unit 25 (see Fig. 4 ) and / or a statistical unit 26 (see Fig. 4 ) are preferably also included in the server 100, but can also be included in the dishwasher 1.

[0095] This arrangement has the particular advantage that for the determination of the dirt types SA, SB, SC (see Fig. 1 ) the high computing power of the server 100 is available, which shortens the investigation time.

[0096] Fig. 6 shows a schematic block diagram of an embodiment of a method for operating a dishwasher 1, in particular the household dishwasher 1 of Fig. 1 or 4 , but also the dishwasher 1 in the arrangement with the computing unit 100 of the Fig. 5. The dishwasher 1 comprises a control device 15 (see Fig. 1 or 4 ) to carry out a washing program for washing soiled items 30 (see Fig. 1 ). In a first step S1, a dirt type SA, SB, SC adhering to the soiled items 30 (see Fig. 1 ) a sensor signal indicative of a plurality of dirt types SA, SB, SC is received. In a second step S2, the dirt type SA, SB, SC adhering to the soiled washware 30 is determined based on the sensor signal. In a third step S3, the wash program is executed based on the determined dirt type SA, SB, SC.

[0097] Although the present invention has been described using exemplary embodiments, it can be modified in many ways. Reference symbols used:

[0098] 1 Dishwasher 2 Washing container 3 Door 4 Washing chamber 5 Swivel axis 6 Loading opening 7 Floor 8 Ceiling 9 Rear wall 10 Side wall 11 Side wall 12 Dishware holder 13 Dishware holder 14 Dishware holder 15 Control device 16 Receiving unit 17 Camera 18 User interface 20 Determination unit 21 Image processing unit 22 Database unit 23 Generation unit 24 Comparison unit 25 Learning unit 26 Statistics unit 27 Communication unit 30 Dishware 50 External unit 100 Computing unit a_11 Dirt parameter a_12 Dirt parameter a_1m Dirt parameter a_21 Dirt parameter a_n1 Dirt parameter a_nm Dirt parameter A Extraction direction DATA Communication connection E Insertion direction MXAMatrix MXBMatrix MXCMatrix MXiMatrix S1 Process step S2 Process step S3 Process step SA Dirt type SB Dirt type SC Dirt type S MXMatrix

Claims

1. Dishwasher (1), in particular a household dishwasher, having a controller (15) for executing a washing program for washing dirty items to be washed (30), having a receiving unit (16) for receiving a sensor signal indicative of a dirt type (SA, SB, SC) adhering to the dirty items to be washed (30) from a plurality of dirt types (SA, SB, SC), wherein each of the dirt types (SA, SB, SC) is determined by a dirt parameter matrix (MXA, MXB, MXC, SMX), in which each position (a_11 - a_nm) is assigned to a dirt parameter, wherein a determining unit (20) is provided, said determining unit being configured to determine the dirt type (SA, SB, SC) adhering to the dirty items to be washed (30) as a function of the received sensor signal, wherein the controller (15) is configured to execute the washing program as a function of the determined dirt type (SA, SB, SC), wherein the sensor signal comprises image information of the dirty items to be washed (30), wherein the determining unit (20) comprises an image processing unit (21) which is configured to determine a value of at least one of the dirt parameters (a_11 - a_nm) contained in the dirt parameter matrix (MXA, MXB, MXC, SMX) as a function of the image information, wherein a database unit (22), a generating unit (23) and a comparison unit (24) are provided, wherein the database unit (22) is configured to store a dirt parameter matrix (MXA, MXB, MXC) for each of the plurality of dirt types (SA, SB, SC), wherein one respective dirt parameter matrix (MXA, MXB, MXC) comprises a predetermined value or value range for each of the dirt parameters (a_11 - a_nm), wherein the generating unit (23) is configured to generate a current dirt parameter matrix (MXi) as a function of the determined value of the at least one dirt parameter (a_11 - a_nm), wherein the comparison unit (24) is configured to compare the generated current dirt parameter matrix (MXi) with at least one subset of the dirt parameter matrices (MXA, MXB, MXC) stored in the database unit (22) and to output a comparison result, and wherein the determining unit (20) is configured to determine the dirt type (SA, SB, SC) adhering to the dirty items to be washed (30) as a function of the comparison result.

2. Dishwasher according to claim 1, characterised in that a learning unit (25) is provided, said learning unit being configured to adapt at least one of the dirt parameter matrices (MXA, MXB, MXC) stored in the database unit (22) as a function of determined values of at least one dirt parameter (a_11 - a_nm).

3. Dishwasher according to one of claims 1 or 2, characterised in that the sensor signal comprises information relating to a washing liquor used for washing the dirty items to be washed (30).

4. Dishwasher according to one of claims 1 to 3, characterised in that the sensor signal comprises information relating to a food prepared and / or consumed by a user of the dishwasher (1) in a specific time interval before a start of the washing program.

5. Dishwasher according to one of claims 1 to 4, characterised in that the controller (15) is configured to adapt the execution of a running washing program.

6. Dishwasher according to one of claims 1 to 5, characterised in that a camera (17) is arranged on the dishwasher (1), wherein the camera (17) is configured to detect an optical sensor signal of the items to be washed (40) and to output the detected optical sensor signal to the receiving unit (16).

7. Dishwasher according to one of claims 1 to 6, characterised in that a statistics unit (26) is provided, said statistics unit being configured to determine a frequency distribution which comprises a frequency of the occurrence of different dirt types (SA, SB, SC), wherein the determining unit (20) is configured to determine the dirt type (SA, SB, SC) adhering to the dirty items to be washed as a function of the frequency distribution.

8. Dishwasher according to one of claims 1 to 7, characterised in that a communication unit (27) is provided, said communication unit being configured to communicate with an external unit (50), in particular with an external kitchen appliance, a mobile device and / or a server.

9. Dishwasher according to one of claims 1 to 8, characterised in that a user interface (18) is provided for detecting a user input as a function of a determined dirt type (SA, SB, SC) by the determining unit (20).

10. Arrangement having a dishwasher (1), in particular a household dishwasher, and a computing unit (100), wherein the dishwasher (1) and the computing unit (100) are configured for data communication (DATA) with one another, wherein the dishwasher (1) comprises a controller (15) for executing a washing program for washing dirty items to be washed (30) and the computing unit (100) comprises a receiving unit (116) for receiving a sensor signal indicative of a dirt type (SA, SB, SC) adhering to the dirty items to be washed (30) from a plurality of dirt types (SA, SB, SC), wherein each of the dirt types (SA, SB, SC) is determined by a dirt parameter matrix (MXA, MXB, MXC, SMX) in which each position (a_11 - a_nm) is assigned to a dirt parameter, wherein a determining unit (120) is provided, said determining unit being configured to determine the dirt type (SA, SB, SC) adhering to the dirty items to be washed (30) as a function of the received sensor signal, and wherein the controller (15) is configured to execute the washing program as a function of the determined dirt type (SA, SB, SC), wherein the sensor signal comprises image information of the dirty items to be washed (30), wherein the determining unit (20) comprises an image processing unit (21) which is configured to determine a value of at least one of the dirt parameters (a_11 - a_nm) contained in the dirt parameter matrix (MXA, MXB, MXC, SMX) as a function of the image information, wherein a database unit (22), a generating unit (23) and a comparison unit (24) are provided, wherein the database unit (22) is configured to store a dirt parameter matrix (MXA, MXB, MXC) for each of the plurality of dirt types (SA, SB, SC), wherein one respective dirt parameter matrix (MXA, MXB, MXC) comprises a predetermined value or value range for each of the dirt parameters (a_11 - a_nm), wherein the generating unit (23) is configured to generate a current dirt parameter matrix (MXi) as a function of the determined value of the at least one dirt parameter (a_11 - a_nm), wherein the comparison unit (24) is configured to compare the generated current dirt parameter matrix (MXi) with at least one subset of the dirt parameter matrices (MXA, MXB, MXC) stored in the database unit (22) and to output a comparison result, and wherein the determining unit (20) is configured to determine the dirt type (SA, SB, SC) adhering to the dirty items to be washed (30) as a function of the comparison result.

11. Method for operating a dishwasher (1), in particular a household dishwasher, having a controller (15) for executing a washing program for washing dirty items to be washed (30), wherein the dishwasher is embodied in accordance with claim 1, the method comprising: • receiving (S1) a sensor signal indicative of a dirt type (SA, SB, SC) adhering to the dirty items to be washed (30) from a plurality of dirt types (SA, SB, SC), wherein each of the dirt types (SA, SB, SC) is determined by a dirt parameter matrix (MXA, MXB, MXC, SMX) in which each position (a_11 - a_nm) is assigned to a dirt parameter, • determining (S2) the dirt type (SA, SB, SC) adhering to the dirty items to be washed (30) as a function of the sensor signal, • executing (S3) the washing program as a function of the determined dirt type (SA, SB, SC).

12. Computer program product which comprises commands which, when the program is executed by a computer, cause the computer to operate the dishwasher according to claim 1 in accordance with the method according to claim 11.