DE102012204229A1 - Cooking appliance device e.g. baking oven, for cooking food, has recognition unit recognizing food in cooking chamber, and selection unit selecting determined cooking programs from predetermined cooking programs in dependence of image - Google Patents

Abstract

The device has a cooking chamber (2) for cooking food, and a controlling unit (3) carrying out predetermined cooking programs. A camera (4) is arranged in the cooking chamber for recording an image (B) of the food in the cooking chamber. A recognition unit (5) recognizes the food provided in the cooking chamber in dependence of the captured image, and a selection unit (6) selects the determined cooking programs from predetermined cooking programs in dependence of the captured image. The recognition unit determines membership probability for certain class in dependence of the captured image. An independent claim is also included for a method for operating a cooking appliance.

Description

The present invention relates to a device for a cooking appliance and a cooking appliance, such as an oven, a microwave or a steamer. Furthermore, the invention relates to a method for operating a cooking appliance.

A conventional cooking appliance has a cooking chamber for cooking food, a control device for carrying out a plurality of predetermined cooking programs and a user interface coupled to the control device for actuating the control device. The user can select a specific cooking program via the user interface. The control device then executes the selected cooking program.

Consequently, the user manually selects a cooking program corresponding to the food to be cooked in the cooking space via the user interface and starts the execution of the cooking program.

Against this background, an object of the present invention is to improve the selection of the appropriate cooking program for a cooking appliance.

According to the invention, a device for a cooking appliance is proposed which has a cooking chamber for cooking food, a control device for carrying out a plurality of predetermined cooking programs, at least one camera arranged in the cooking chamber for taking at least one image of a cooking product located in the cooking chamber, a detection means for Recognizing the arranged in the cooking space Garguts depending on the at least one recorded image, and a selection means for selecting a particular cooking program from the plurality of predetermined cooking programs in dependence on the detected food.

Depending on the food in the cooking chamber can be selected with the help of one or more cameras, which are mounted in the oven, the appropriate cooking program. Thus, the cooking appliance can actively recognize which program or cooking program is suitable for the food. Thus, the user has taken the work to search the appropriate cooking program only from a list of possible cooking programs. In addition, the automatic selection of the cooking program prevents a user who does not know all possible cooking programs from finding the appropriate cooking program due to his ignorance.

In one embodiment, the detection means is adapted to associate the food in the cooking chamber in dependence on the at least one recorded image of a particular class of a plurality of predetermined classes for food items. The respective class is determined by a class pattern specific to the class.

The classifications and the associated classes of food can make the probabilities of affiliation softer.

In a further embodiment, the recognition means is further configured to determine a membership probability for the particular class in dependence on the at least one captured image.

Thus, the cooking appliance recognizes with the help of the at least one camera and reference values, in particular the class patterns, with a certain probability a food in the cooking chamber. Depending on the algorithm, a program can be passively selected and executed in the background according to the recognition or actively proposed to the user by means of a user interface.

In another embodiment, the recognition means is adapted to extract certain features from the at least one captured image and to classify the food in the cooking space by comparing the extracted specific features with the class patterns.

In a further embodiment, the control device is configured to execute the cooking program determined by the selection means if the determined membership probability is greater than a predetermined threshold value.

In a further embodiment, a user interface coupled to the control device is provided for activating the control device. The control device is preferably configured to output the cooking program determined by the selection means to a user by means of the user interface for execution if the determined membership probability is greater than a predetermined threshold value. This allows the user to choose the particular cooking program to run.

In a further embodiment, a user interface coupled to the control device is provided for activating the control device, wherein the control device is set up to output the cooking program determined by the selection means to a user via the user interface for execution, if the determined membership probability is greater than a lower threshold and less than or equal to an upper threshold, or directly execute if the determined membership probability is greater than the upper threshold.

Thus, the active and passive selection of the cooking program can be adjusted by means of the thresholds and their choice.

By means of the threshold values, the safety of the selection of the suitable cooking program can be set and / or maximized.

In a further embodiment, the respective class pattern for the respective class comprises at least one set of reference values for the food of this class.

In another embodiment, the at least one set of reference values comprises a first set of reference values for the geometric shape of the food of this class, a second set of reference values for the color of the food of that class, and a third set of reference values for the surface quality of the food of that class ,

In a further embodiment, the detection means uses a SURF algorithm for detecting the food arranged in the cooking chamber. SURF is an algorithm by Herbert Bay et al. for fast and robust recognition of image features for machine vision ( Herbert Bay, Tinne Tuytelaars, and Luc Van Gool: SURF: Speeded Up Robust Features, Proceedings of the 9th European Conference on Computer Vision, Springer Publishing, 2006 ).

Furthermore, it is possible to use a combination, for example a weighted combination, of different algorithms for object recognition of the food in the cooking chamber.

With the SURF algorithm, so-called "interest points" can be extracted from the recorded image as features. The extracted interest points can then be compared with those of another image, in particular a reference image, in order to draw conclusions regarding the similarity of the two images. These comparison images can be part of the class patterns.

In another embodiment, the detection means is adapted to learn from previous detections of arranged in the cooking chamber food.

For example, the cooking appliance can store the reference values of the cooking image when starting a cooking program and potentially transmit them to the server. As a result, the cooking appliance recognizes the next time the same food, especially if it is similarly positioned, again. For the learning process, the positioning of a second, opposite camera is advantageous in order to receive the food from two sides and thus to extract reference values from both sides.

In a further embodiment, the control device of the cooking appliance integrates the recognition means and the selection means.

In a further embodiment, the selection means is integrated in a server which is coupled via a network to the oven, in particular to the detection means and the control device of the oven.

Accordingly, if the cooking appliance has the possibility to access a network, for example the Internet, then the cooking appliance can transmit the acquired reference values from the recorded image of the food currently in the cooking chamber to the server and in response a concrete cooking program or alternatively a program category get a certain probability.

The respective means, recognition means and selection means can be implemented in hardware and / or software technology. In a hardware implementation, the respective means may be designed as a device or as part of a device, for example as a computer or as a microprocessor. In a software implementation, the respective means may be designed as a computer program product, as a function, as a routine, as part of a program code or as an executable object.

Furthermore, a cooking appliance is proposed which has a device as described above. The cooking appliance is for example an oven, a microwave, a steamer or a warming drawer.

Furthermore, a method for operating a cooking appliance with a cooking chamber for cooking food and a control device for performing a plurality of predetermined cooking programs in the oven is proposed. In this case, at least one image of a cooking product located in the cooking chamber is received by means of at least one camera arranged in the cooking chamber. Then, the cooking product arranged in the cooking chamber is detected as a function of the at least one recorded image. Consequently, a specific cooking program is selected from the plurality of predetermined cooking programs depending on the detected food.

Furthermore, a computer program product is proposed, which causes the execution of at least one step of a method for operating a cooking appliance, as explained above, on a program-controlled device.

A computer program product such as a computer program means can be provided or supplied, 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 communication network by the transmission of a corresponding file with the computer program product or the computer program means.

Further advantageous embodiments and aspects of the invention are the subject of the dependent claims and the embodiments of the invention described below. Furthermore, the invention will be explained in more detail by means of preferred embodiments with reference to the attached figures.

It shows:

1 a schematic view of a first embodiment of a cooking appliance;

2 a schematic view of a second embodiment of a cooking appliance;

3 a schematic view of a third embodiment of a cooking appliance;

4 a schematic view of a fourth embodiment of a cooking appliance, which is coupled to a server; and

5 a schematic flow diagram of an embodiment of a method for operating a cooking appliance.

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

In 1 is a schematic view of a first embodiment of a cooking appliance 1 shown. The cooking appliance 1 has a cooking space 2 for cooking food (not shown). For example, the cooking appliance 1 designed as an oven. In this embodiment, the cooking chamber 2 then be referred to as a stovepipe.

Furthermore, the cooking appliance has 1 a control device 3 for performing a plurality of predetermined cooking programs for cooking in the oven 2 located Garguts. The control device 3 integrates a detection means 5 and a selection tool 6 and is designed, for example, as a microcontroller. For this purpose, the control device 3 control the cooking chamber by means of at least one control signal S1.

In the oven 2 is at least a camera 4 arranged, which is adapted to a picture B of the in the cooking chamber 2 to absorb the food to be cooked. The captured image B can be taken from the camera 4 to the control device 3 , in particular to the detection means 5 , be transmitted. Likewise, it is possible to have a plurality of cameras 4 in the oven 2 which then a plurality of images B, in particular from different views, to the detection means 5 transfer.

The detection means 5 recognizes that in the oven 2 arranged food in dependence on at least one recorded image B. It sets the detection means 5 an image recognition algorithm, such as the SURF algorithm, for detecting the in the cooking chamber 2 arranged Garguts.

Here is the detection means 5 preferably adapted to that in the oven 2 Depending on the at least one recorded image B to be located Gargut associated with a particular class of a plurality of predetermined classes for food. Examples of such classes of food are poultry, pork, beef and vegetables. The respective class is determined by a class pattern specific to the class.

Depending on the at least one recorded image B, the recognition means 5 also be set to determine a membership probability for the particular class. For the example of four such classes for pork, beef, poultry and vegetables an exemplary distribution could be: 91% for pork, 6% for beef, 2% for poultry and 1% for vegetables.

The respective class pattern for the respective class comprises at least one set of reference values for the food of this class. The at least one set of reference values may comprise a first set of reference values for the geometric shape of the food of this class, a second set of reference values for the color of the food of that class, and a third set of reference values for the surface quality of the food of that class. Other sets of reference values for a particular class are conceivable. Furthermore, the recognition means 5 be adapted to extract certain features from the at least one captured image and in the oven 2 To classify food to be located by means of a comparison of the extracted specific features with the class patterns. Depending on the classification then the selection means 6 select the suitable cooking program P1-P3.

With regard to the execution of the selected cooking program P1-P3 several alternatives are possible. These several alternatives include in particular

following:

According to a first alternative, the control device 3 be adapted to that of the selection means 6 certain cooking program P1-P3 automatically execute, if the certain membership probability is greater than a predetermined threshold, for example 90%.

According to a second embodiment, a user is offered the selected cooking program P1-P3 by means of a user interface. Details are related to 4 described.

Furthermore, the recognition means 5 to be set up from previous detections in the cooking chamber 2 to learn arranged food. In this learning process, in particular, the threshold values and the decision making depending on the at least one recorded image B can be adjusted.

In the embodiments of the 1 to 3 integrates the control device 3 of the cooking appliance 1 the detection means 5 and the selection tool 6 , An alternative to this is with reference to 4 shown.

In 2 is a schematic view of a second embodiment of a cooking appliance 1 shown. The second embodiment of the 2 is based on the first embodiments of 1 and has all the features of the first embodiment of 1 on. In addition, the control device has 3 of the 2 a storage device 7 , The storage device 7 For example, in the control device 3 be integrated. Alternatively, the storage device 7 with the control device 3 be coupled. The storage device 7 stores the predetermined cooking programs P1-P3. For example, the storage device 7 be configured as read-only memory, for example as ROM. The storage device 7 can also be designed as a memory device which can be modified externally by the user. That is, the user can then new or updated cooking programs in the storage device 7 import.

3 shows a third embodiment of a cooking appliance 1 , The third embodiment of 3 is based on the second embodiment of the 2 and has all the features of the second embodiment. In addition, the cooking appliance has 1 of the 3 a user interface 8th for driving the control device 3 , The user interface 8th can have a touch-sensitive screen (touchscreen and / or control buttons). In the embodiment of 3 represents the selection means 6 the specific cooking program P1-P3 to the user via the user interface 3 out. The user may then select the illustrated particular cooking programs P1-P3 for execution. For this purpose, a second control signal S2 from the user interface 8th to the control device 3 be transmitted. Preferably, that of the selection means 6 certain cooking program P1-P3 is output to the user only if the determined membership probability is greater than a predetermined threshold.

Alternatively, that of the selection means 6 certain cooking program P1-P3 to the user via the user interface 8th for selection and execution, when the determined membership probability is greater than a lower threshold and less than or equal to an upper threshold. However, if the certain membership probability is greater than the upper threshold value, the control device 3 execute the specific cooking program P1-P3 directly without prior output to the user.

In 4 is a fourth embodiment of a cooking appliance 1 shown. The cooking appliance 1 of the 4 differs from the embodiment of 3 in that the selection means 6 not in the cooking appliance 1 is integrated, but in a server 9 , The server 9 is over a network 10 For example, the Internet, with the cooking appliance 1 coupled. The server 9 For example, from the manufacturer of the cooking appliance 1 operated. In the embodiment of 4 can be the detection agent 5 a request A, which includes the recognized food and preferably the associated membership probability, to the server 9 and thus to the selection means 6 be transmitted. The selection tool 6 then selects the appropriate cooking program P1-P3 and transmits this selected cooking program P1-P3 to the control device 3 which can then execute this cooking program P1-P3. Alternatively, the selection means 6 even just a command to the controller 3 which indicates which of the specific cooking programs P1-P3 is to be executed.

In 5 is a schematic flow diagram of an embodiment of a method for operating a cooking appliance 1 shown. The Cooking appliance 1 has a cooking space 2 for cooking food and a control device 3 for performing a plurality of predetermined cooking programs P1-P3.

In step 501 At least one image B of that in the cooking chamber 2 located food by means of at least one in the oven 2 arranged camera 4 added.

In step 502 that will be in the oven 2 arranged food detected as a function of at least one recorded image B by means of an object detection.

In step 503 a specific cooking program P1-P3 selected from the plurality of predetermined cooking programs P1-P3 depending on the detected food and in particular executed.

Although the present invention has been described with reference to embodiments, it is variously modifiable.

LIST OF REFERENCE NUMBERS

1

 Cooking appliance

2

 oven

3

 control device

4

 camera

5

 recognition means

6

 selection means

7

 storage device

8th

 User interface

9

 server

10

 network

501-503

 step

A

 inquiry

B

 image

P1

 first cooking program

P2

 second cooking program

P3

 third cooking program

S1

 first control signal

S2

 second control signal

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited non-patent literature

• Herbert Bay, Tinne Tuytelaars, and Luc Van Gool: SURF: Speeded Up Robust Features, Proceedings of the 9th European Conference on Computer Vision, Springer Verlag, 2006 [0019]

Claims (15)

Device for a cooking appliance ( 1 ), with a cooking space ( 2 ) for cooking food, a control device ( 3 ) for performing a plurality of predetermined cooking programs, at least one in the cooking chamber ( 2 ) arranged camera ( 4 ) for picking up at least one image (B) of one in the cooking space ( 2 ), a detection means ( 5 ) for detecting the in the cooking space ( 2 ) arranged Garguts depending on the at least one recorded image (B), and a selection means ( 6 ) for selecting a specific cooking program (P1-P3) from the plurality of predetermined cooking programs (P1-P3) in dependence on the detected food. Device according to claim 1, characterized in that the recognition means ( 5 ) is set up in the cooking chamber ( 2 ) to be associated Gargut depending on the at least one recorded image (B) of a particular class of a plurality of predetermined classes for food, the respective class is determined by a class pattern. Device according to claim 2, characterized in that the recognition means ( 5 ) is further adapted to determine a membership probability for the particular class in dependence on the at least one captured image (B). Device according to claim 2 or 3, characterized in that the recognition means ( 5 ) is adapted to extract certain features from the at least one recorded image (B) and in the cooking chamber ( 2 ) to be classified by means of a comparison of the extracted certain features with the class patterns. Device according to claim 4, characterized in that the control device ( 3 ) is set up by the selection means ( 6 ) execute certain cooking program (P1-P3) if the determined membership probability is greater than a predetermined threshold. Apparatus according to claim 4, characterized in that a with the control device ( 3 ) coupled user interface ( 8th ) for driving the control device ( 3 ) is provided, wherein the control device ( 3 ) is set up by the selection means ( 6 ) certain cooking program (P1-P3) a user via the user interface ( 8th ) for execution if the determined membership probability is greater than a predetermined threshold. Apparatus according to claim 4, characterized in that a with the control device ( 3 ) coupled user interface ( 8th ) for driving the control device ( 3 ) is provided, wherein the control device ( 3 ) is set up by the selection means ( 6 ) certain cooking program (P1-P3) a user via the user interface ( 8th ) for execution if the determined membership probability is greater than a lower threshold and less than or equal to an upper threshold, or direct execution if the determined membership probability is greater than the upper threshold. Device according to one of claims 2 to 7, characterized in that the respective class pattern for the respective class comprises at least one set of reference values for the food of this class. Apparatus according to claim 8, characterized in that the at least one set of reference values comprises a first set of reference values for the geometric shape of the food of this class, a second set of reference values for the color of the food of that class and a third set of reference values for the surface finish of the food of this class. Device according to one of claims 1 to 9, characterized in that the detection means ( 5 ) a SURF algorithm for detecting the in the cooking chamber ( 2 ) arranged Garguts uses. Device according to one of claims 1 to 10, characterized in that the detection means ( 5 ) is adapted from previous detections in the cooking chamber ( 2 ) to learn arranged food, Device according to one of claims 1 to 11, characterized in that the control device of the cooking appliance ( 1 ) the recognition means ( 5 ) and the selection means ( 6 ) integrated. Device according to one of claims 1 to 11, characterized in that the selection means ( 6 ) in a server ( 9 ) which is connected via a network ( 10 ) with the cooking appliance ( 1 ), in particular with the detection means and the control device of the oven is coupled. Cooking appliance ( 1 ) with a device according to one of claims 1 to 12. Method for operating a cooking appliance ( 1 ) with a cooking space ( 2 ) for cooking food and a control device ( 3 ) for performing a plurality of predetermined cooking programs (P1-P3), comprising the steps of: Take up ( 501 ) at least one image (B) of one in the cooking chamber ( 2 ) located by means of at least one in the cooking space ( 2 ) arranged camera ( 4 ), Detect ( 502 ) of the cooking product arranged in the cooking chamber as a function of the at least one recorded image, and selecting ( 503 ) of a certain cooking program (P1-P3) of the plurality of predetermined cooking programs (P1-P3) in dependence on the detected food.

Images