A method of controlling a cooking appliance, and a cooking appliance for carrying out said method

Abstract

The present invention relates to a method for handling a cooking appliance, in particular an oven, that allows a user to control and affect the browning level of the food to be cooked. The invention also pertains to a cooking appliance for implementing the method. The process of sensing the browning level of the food to be cooked comprises, at regular time intervals, optimal lighting, for instance by means of a lighting means (6), and sensing, for instance by means of a device (7) to capture and store images and / or videos inside the cooking appliance, and, during the entire cooking session, a precise recognition of the browning level of the food to be cooked by means of a model (IV). The system for sensing the browning level of the food to be cooked comprises without limitations a cooking appliance (I), for instance an oven coupled with a processing unit (II), storage (III), model (IV), and a post-processing unit (V), wherein at least one or all of the devices (II, III, IV, V) can be integrated in the cooking appliance (I) or in a separate external device, for instance a personal computer, mobile device, cloud storage, and similar.

Description

A METHOD OF CONTROLLING A COOKING APPLIANCE, AND A COOKING APPLIANCE FOR CARRYING OUT SAID METHOD

[0001] The present invention relates to a method for handling a cooking appliance, in particular an oven with which a user may monitor and control a browning level of the food to be cooked. The invention also relates to the cooking appliance for carrying out said method.

[0002] The method and cooking appliance, in particular the oven referred to above, are disclosed in document CN 106175478 A. Described therein is an oven comprising a cooking chamber formed to receive the food to be cooked, a user interface formed to display information related to a cooking procedure, a first and second energy source, and cooking controls. The said controls, which are coupled with the energy sources, can comprise a processing circuit board designed in such a way that it allows the user to select the browning level of the food to be cooked via the user interface. The user forwards instructions to a selected control console displayed by the user interface. Selecting the browning level of the food to be cooked ensures the control parameters for directing heat toward the food to be cooked by means of the second energy source. The shortcomings of the said known solution are that it requires two energy sources, leading to higher energy consumption and a complex handling algorithm.

[0003] Furthermore, the company BSH offers a method and cooking appliances, including the Siemens iQ700 oven and the Bosch Series 8 HNG978QB1A oven. These two ovens are equipped with a built-in camera or a sensor unit, which is arranged in the upper left corner of the oven, and two LED lighting devices, which are arranged on the side walls of the cooking chamber. To affect the browning level of food, the user can select the browning level on a five-point scale either on the oven display or by means of a mobile application. As the process starts, an initial photograph of the food in the oven is taken, after which the food inside the cooking chamber cannot be moved or changed in any way. Once cooking begins, the camera continuously monitors the food, capturing an image every 10 seconds. Each captured image is tagged with its browning level and compared to the browning level selected by the user at the start of the process. The cooking appliance turns off automaticallyonce the desired browning level has been reached. A shortcoming of the said solution is the camera’s placement in the upper corner, which results in an unnatural view of the food and makes recognition less effective. Another significant issue is the 10-second interval for monitoring, as the cooking process is not as fast.

[0004] The object of the present invention is to create a method for handling a cooking appliance, in particular an oven, with which a user can monitor and control the browning level of the food to be cooked, which would address the shortcomings of the known solutions.

[0005] The object according to the present invention is accomplished by the features set forth in the characterising portion of claim 1. The details of the invention are disclosed in the corresponding sub-claims.

[0006] The invention is further described in detail by way of a non-limiting embodiment and with a reference to the accompanying drawings whereFig. 1 shows a schematic cross-sectional view of a cooking appliance for carrying out a method in the first vertical plane,Fig. 2 shows a schematic cross-sectional view of the cooking appliance of Fig. 1 in the second vertical plane,Fig. 3 shows a schematic cross-sectional view of the preferred embodiment of the cooking appliance for carrying out a method in the first vertical plane,Fig. 4 shows a schematic view of the preferred embodiment of the cooking appliance of Fig. 3 as seen from above,Fig. 5 shows a lighting means,Fig. 6 shows the lighting means as seen from the side,Fig. 7 shows the lighting means as seen from the front,Fig. 8 shows the lighting means in a three-dimensional view, andFig. 9 shows a schematic view of a system for sensing the browning level.

[0007] A cooking appliance according to the invention comprises a cooking chamber 1 composed of walls 2, which can be closed by means of a door 3. The cooking chamber 1 contains, in the horizontal plane, at least one tray 4 for accommodating food, which is to be thermally treated.

[0008] Arranged in the area above the upper wall 5 of the walls 2 are at least one lighting means 6 for illuminating the tray 4 for accommodating food, which is to be thermally treated, and for illuminating the interior of the cooking chamber 1, and at least one device 7 to capture and store images and / or videos. The entry of a light beam from the lighting means 6 into the interior of the cooking chamber 1 and the capture of images and / or videos with the said device 7 to capture images and / or videos is enabled through at least one opening in the upper wall 5 of the walls 2. Furthermore, in and / or on at least one side wall 8; 9 of the walls 2, at least one additional lighting means 10 is provided to illuminate the interior of the cooking chamber 1. For the device 7 to capture images and / or videos to be able to capture an image and / or video, at least one lighting means 6; 10 must be turned on.

[0009] Also possible is an embodiment of the cooking appliance where at least one said device 7 to capture and store images and / or videos is arranged in the area above the door 3 in the front wall of the walls 2.

[0010] At least one of the lighting means 6; 10 is, for example, selected as a light-emitting diode (LED) and at least one of the lighting means 6; 10 is, for example, selected as a light guide rod 6a.

[0011] Furthermore, at least one device 7 to capture and store images and / or videos is, for example, selected as a digital camera to capture images and / or videos in high resolution, for instance 1920x1080 pixels or more.

[0012] The lighting means 6 comprises without limitations a light guide rod 6a to which at first end thereof arranged outside the cooking chamber 1 there is attached a housing 6b,which is provided on the side facing away from the light guide rod 6a with a heat sink 6c. Inside the housing 6b, a light-emitting module 6d is arranged, which is in a light connection with the light guide rod 6a. The lighting means 6, which with a free end 6e of the light guide rod 6a extends in the interior of the cooking chamber 1, can be arranged anywhere in the upper area of the cooking appliance or walls 2, above the highest tray 4 level. Particularly is preferred such an arrangement of the lighting means 6 in relation to the device 7 to capture and store images and / or videos that the entire tray 4 is illuminated without obstructing the view of the device 7 to capture and store images and / or videos, wherein no light effects are generated, such as reflection, glare, or backlight, that would affect the quality of the image and / or video.

[0013] The lighting means 6 can be, in the horizontal plane along the centre line from the door 3 toward the back wall of the cooking chamber 1, arranged behind or in front of the device 7 to capture and store images and / or videos. To ensure optimal lighting, the lighting means 6 and the device 7 have approximately identical orientations of their longitudinal centre lines.

[0014] According to an alternative embodiment, it is provided that the lighting means 6 and the device 7 to capture and store images and / or videos are arranged next to each other on the horizontal plane in the longitudinal direction of the front wall of the walls 2.

[0015] The device 7 for capturing and storing images and / or videos is generally arranged anywhere in the upper area of the cooking appliance or walls 2, above the highest tray 4 level. The preferable arrangement of the device 7 to capture and store images and / or videos provides for that it is arranged in the upper horizontal plane in the area of the door 3, which ensures a capture of image and / or video from a perspective that is best for the user. Also possible is an embodiment in which the device 7 is arranged behind the door 3 or on any other wall of the walls 2, above the highest tray 4 level.

[0016] It is provided for, according to the preferred embodiment, that the lighting means 6 and the device 7 to capture and store images and / or videos and, respectively, the longitudinal centre lines of the lighting means 6 and the device 7, are, in relation to thehorizontal plane of the cooking appliance, tilted at an angle a in the direction from the back wall of the walls 2 toward the door 3, in such a manner that the ends of the lighting means 6 and light guide rod 6a, which extend or protrude into the cooking chamber 2, and the device 7 face the back wall of the walls 2. The said angle a of the light guide rod 6a and the device 7 is preferably identical and falls within a range of approximately 0° to approximately 45°. Also possible is an embodiment where the angle a of the inclination of the light guide rod 6a falls within a range of approximately 0° to approximately 45°, while the angle of the inclination of the device 7 is different, either smaller or larger.

[0017] With the preferred embodiment of the cooking appliance, the free end 6e of the light guide rod 6a, i.e. the end of the light guide rod 6a that protrudes into the cooking chamber 1, is formed with an inclination 11 comprising an angle (3 in relation to the longitudinal centre line of the lighting means 6. The said angle (3 falls within a range of approximately 30° to approximately 60°. Furthermore, the lighting means 6 or the light guide rod 6a is rotated around the longitudinal centre line thereof in such a manner that the normal of the inclination 11 extends at least approximately in the direction toward the relatively narrow area of the closest bottom corner of the cooking chamber 1. It is obvious that the angular rotation of the light guide rod 6a around its own longitudinal axis depends on the dimensions of the cooking chamber 1. However, the light guide rod 6a is rotated around its longitudinal centre line in such a manner that the majority of the light exiting the free end 6e faces the area of the most distant corner of the cooking chamber 1. This means that the radiation light flux exiting through the free end 6e of the light guide rod 6a, faces the area of the most distant corner in such a way that the said area is illuminated as much as possible.

[0018] The said inclination 11 is finished in such a manner that it at least partly transmits the light along the light guide rod 6a, wherein the said inclination 11 is at least partly formed as a translucent surface, for instance as a satinised, matt, or frosted surface.

[0019] The said free end 6e of the lighting means 6 or light guide rod 6a, on the side diametrically opposite to the inclination 11, further comprises a surface 12 finished in such a manner that it at least partly transmits the light along the light guide rod 6a. The said surface 12 is, for example, formed as a translucent surface, for instance a satinised, matt, or frostedsurface that, along the perimeter of the light guide rod 6a, extends for at least 180° and symmetrically in relation to the surface 12. Furthermore, the said surface 12, in the longitudinal direction of the light guide rod 6a, extends from the tip 11’ of the inclination 11 for a length I, which corresponds to at least the longest dimension of the surface 12 according to the longitudinal centre line of the light guide rod 6a.

[0020] Also possible is an embodiment where the light guide rod 6a is transparent, while the free end 6e of the rod is translucent. A translucent design of the free end 6e reduces the luminosity of the lighting means 6 and more evenly disperses light inside the cooking chamber 1.

[0021] The free end 6e of the light guide rod 6a can be formed in different ways, including rectilinearly along the longitudinal centre line of the light guide rod 6a or curved out of the direction of the longitudinal centre line of the light guide rod 6a, for example in the shape of the letter L. The possible, yet non-limiting shapes of the free end 6e where light exits, are convex, concave, chamfered on one side or multiple sides, or rounded.

[0022] To prevent an unwanted dispersion of light from the light guide rod 6a, the latter can be at least partly surrounded by a mantle, which is at least partly light impermeable.

[0023] According to one of the possible, but non-limiting embodiments, it is provided that the lighting means 6 is formed to emit light with a temperature within the range of approximately 2000 K to 7000 K.

[0024] According to another embodiment, it is provided that the lighting means 6 and the device 7 to capture and store images and / or videos are arranged in parallel or in series in close vicinity next to each other.

[0025] According to yet another embodiment, it is provided that the lighting means 6 is arranged on / in the device 7 to capture and store images and / or videos in such a manner that it at least partly surrounds the lens of the device 7, wherein it does not hinder the collection of light entering the device 7 through the lens.

[0026] The process of sensing the browning level of the food to be cooked comprises sensing inside a cooking appliance in regular time intervals, for instance by means of a device 7 to capture and store images and / or videos, and optimal lighting during the entire cooking session, for instance by means of a lighting means 6, and a precise recognition of the browning level of the food to be cooked, for instance by means of a model IV and a postprocessing unit V. The system for sensing the browning level of the food to be cooked comprises without limitations a cooking appliance I, for instance an oven coupled with a processing unit II, storage III, model IV, and a post-processing unit V, wherein at least one or all of the units II, III, IV, V can be integrated in the cooking appliance I or in a separate external device, for instance a personal computer, mobile device, cloud storage, and similar.

[0027] In the presented embodiment, the processing unit II comprises at least one storage III, at least one model IV, and at least one post- process! ng unit V, wherein it determines the browning level of the food to be cooked based on the data received from the cooking appliance I.

[0028] The storage III is provided to store data received from the cooking appliance I as well as historical data on recognition, which is created during the process by the processing unit II. Data obtained from the cooking appliance I enables the model IV to utilise this information for immediate recognition or for making subsequent improvements through continuous learning. Continuous learning includes an on-going perfection and updating of the machine learning model as soon as new data is available, which allows for adjustments of the model parameters for improved precision and efficiency. Previously acquired recognition data is retained forfuture reference and use in the post- process! ng unitV.

[0029] At least one said model IV for sensing the browning level of the food to be cooked comprises the following features: i) The weights and spacings of the recognition layer IV2 are set to random values by default; ii) The recognition layer IV2 learns to recognise a large selection of different foods, preferably at least two different foods, which the user can select by means of a suitable interface, for instance a computer, mobile device, oven interface, TV, and similar, wherein thelearning can make use of a cyclic learning rate scheduler or a constant learning rate scheduler, and learning lasts for at least two learning epochs, during which the model IV runs through the entire presented set of learning data; iii) The recognition layer IV2 learns to recognise a specific food, wherein it uses the soft labels technique. The soft labels technique uses browning level probability for a defined input image. The input image can be recognised to contain multiple browning levels simultaneously, wherein each browning level has a different probability. The probability can be determined by means of multiple fuzzy logic approaches with different membership functions, wherein the most appropriate one is the membership function as follows. For the selected browning level, for instance the medium done level, probability 1 is determined on the mid-image of the interval for this level with respect to the entire cooking session. Probability 0 is determined for the first and last mid-image of the interval for the previous and next browning level with respect to the entire cooking session. For other images, probability is determined with respect to the linear distance from the best image. The soft labels technique allows the model IV to form a more versatile understanding of the browning level. The model encourages a better generalisation and reduces the risk of over- fitting. The model is sensitive to slight input changes, which improves its robustness.

[0030] The cooking appliance I further comprises at least one device 7 to capture and store images and / or video, for instance a camera provided to capture the characteristics of the food to be cooked, which is arranged inside the cooking chamber 1, and at least one lighting means 6 to illuminate the inside of the cooking chamber 1. Among other, the said device 7 to capture and store images and / or video comprises at least one sensor, which even at high temperatures reliably measures the characteristics of the food to be cooked. At least one said sensor is, for example, an RGB image sensor for obtaining the visual appearance of the food to be cooked or the characteristics of the latter. Moreover, at least one said sensor can be provided as a sensor of humidity and / or sensor of the oxygen content and / or sensor, which obtains spectral informations for identifying barely noticeable changes in the characteristics of the food to be cooked, which includes, but is not limited to, a spectrophotometer or calorimeter, an infrared thermometer, a thermal camera, and similar.

[0031] The said image sensor is provided to capture images of a set size, for instance at least1280 x 960 pixels, in a set format, for instance in raw YUV or JPEG RGB format, wherein the capture of the images starts at a set start time and then follows in set intervals, for instance each 30 seconds. Image capture can begin in multiple ways, preferably starting simultaneously with cooking process initiated by the user. Alternatively, image capture can begin before the user initiates the cooking process or after it.

[0032] At least one said model IV comprises a series of layers, including but not limited to at least one pre-processing layer IV2and at least one recognition layer IV2, wherein the model, based on input data, identifies the browning level of the food to be cooked inside the cooking appliance I. The pre-processing layer IV2prepares data to be input in the recognition layer IV2, which creates a recognition proposal IV3. The latter can then be directed to the postprocessing unitV for further analysis and processing.

[0033] The post-processing unit V obtains the previous result from the storage III and the recognition proposal IV3from the model IV, wherein it dynamically improves the recognition proposal IV3to obtain the most accurate recognition result V3. More specifically, the postprocessing unit V reduces the impact of isolated errors in image recognition at the overall result of recognition.

[0034] The process of handling the cooking appliance I, in particular an oven with which a user can control and thus affect the browning level of the food to be cooked, starts with the insertion of the food into the oven. Next, the user selects the cooking program, for instance via the interface on the oven itself or a mobile device, personal computer, TV, and similar, wherein the said cooking program comprises among other at least parameters about the type of food, cooking temperature, browning level, and cooking time. If the user is satisfied with the default parameters, the process continues with the immediate start of cooking or a delay in the start of cooking. If the user is not satisfied with the default parameters, the user can change the desired parameter, for instance via the interface on the oven itself or a mobile device, personal computer, TV, and similar. The selected browning level serves as the target browning level, upon which the oven, once the target browning level has been reached, communicates the end of cooking.

[0035] If the user is satisfied with the selection of the desired parameters, the cooking appliance I forwards the relevant data, which include but are not limited to the cooking parameters and the target browning level, to the processing unit II, which stores the said relevant data in the storage III. Based on the said relevant data, the processing unit II selects at least one model IV. The processing unit II starts a new cooking session, which lasts until stopped by the user. The cooking session comprises handling the cooking inside the cooking appliance I along with all respective data. It should be noted that the processing unit II is not limited to controlling just one single cooking session or one single cooking appliance I connected to the said cooking session; conversely, the processing unit II can simultaneously control multiple cooking sessions and / or multiple cooking appliances I.

[0036] Once the cooking appliance I starts cooking based on the assigned parameters of the cooking program, the cooking appliance I sends data to the processing unit II, which stores it in the storage III. The model IV obtains the latest data and processes it in the pre-processing layer IVi, where it prepares it with a subsequent analysis in the recognition layer IV2. The preprocessing of data in the pre-processing layer IV2comprises various techniques depending on the type of data. Hence, for instance, if the data is in image form, it is converted to an acceptable resolution, including but not limited to the resolution of 224 x 224 pixels, before normalisation with regular procedures.

[0037] The prepared data is then transferred to the recognition layer I V2, which by means of a suitable algorithm prepares the recognition proposal IV3. The said recognition layer IV2can comprise any machine learning algorithm designed to recognise browning level. The said algorithm comprises different ordinal regression models, which are coupled with clustering algorithms, Naive Bayes classifiers, support vector methods, and decision trees. Deep learning algorithm has been shown to be the most favourable approach. Within this framework, various types are available, such as artificial neural networks, convolutional neural networks (CNNs), recurrent neural networks, self-organising maps, Boltzmann neural networks, autoencoders, generative models, diffusion models, transformers, LSTMs, and others. Preferable in particular is the convolutional neural network (CNN).

[0038] Various architectures can be used as the recognition layer IV2, such as EfficientNet,ResNet, Vision Transformer, ConvNext, MobileNet, and SqueezeNet, but in order to obtain the optimal results, the preferable architecture of choice is EfficientNet v2 S.

[0039] In the next step, the post-processing unit V obtains the previous browning recognition results from the storage 3 and the recognition proposal IV3from the model IV, after which it dynamically improves the recognition proposal IV3from the model IV and creates the most accurate recognition result V3. This increases the knowledge of previous cooking progressions, which reduces potential isolated errors in image recognition. Hence, for instance, the post-processing unit V corrects the browning level when previous recognition results Vi show a low browning level even though the current recognition proposal IV3wrongly proposes that the food has been burnt. With continuous monitoring of the food in the oven at regular intervals, for instance every 30 seconds, it is highly unlikely that the food will burn within a relatively short time. This is why in such a case, the recognition proposal IV3from the model IV is identified as an error and adjusted accordingly.

[0040] For such an adjustment, it is possible to use different approaches, including but not limited to the following methods:(i) Maximum value method: This method only updates the browning level if the current recognition proposal IV3surpasses the previous recognition result in the storage III;(ii) Window processing method: With this method, the recognition result V3is determined with the averaging of previous recognition results V3by means of an inbound recognition proposal IV3from the model IV;(iii) Score processing method: This method is similar to the maximum value method, but it relies on the probability result, not on the model output. The browning level is changed in the event that the probability surpasses the predefined threshold.

[0041] Once the recognition result V3has been obtained, the processing unit II determines the next steps, which can vary based on the following scenarios:(i) If the recognition result V3shows that the browning level has not been reached yet, the processing unit II decides to continue the cooking session;(ii) If the recognition result V3confirms that the browning level has been reached, the processing unit II can notify the user or automatically end the cooking session. The user caneither continue the cooking or stop it;(iii) If the recognition result Vi proposes that the target browning level has been surpassed, the processing unit II either notifies the user or automatically ends the cooking session.

[0042] Once the cooking in the cooking appliance I ends, recognition with the device 7 to capture and store images and / or video of the food to be cooked ends immediately, while the illumination by means of the lighting means 6 stops with a delay, either immediately or gradually by reducing the intensity to zero. A heat sink 6c remains operational until the oven cools down.

[0043] The browning level of the food to be cooked in the oven of the cooking appliance I according to the invention represents the appearance of food during the entire cooking session, wherein it is specifically determined as a defined range of browning levels. During the cooking session, the colour of the food changes from the least brown through golden brown and lastly to the highest browning level, which is why each food is associated with multiple browning levels:(i) Recommended level: Represents a default setting when the food is cooked right, typically following a recipe, which is reflected in a golden yellow shade;(ii) Optional level: Foresees either the lowest or the highest browning level compared to the recommended level, wherein the user has the flexibility to set the browning level based on the user’s preference. The number of optional levels and their granularity can be changed depending on the type of food;(iii) Ancillary level: Helps the processing unit II make more accurate decisions, which the users cannot directly select themselves. While the number of ancillary levels can vary, there are always at least two distinct levels, namely the level “raw”, which marks the lowest or null browning, and the level “burnt”, which marks the highest browning level;(iv) Target level: Represents the browning level that the user selects for the cooking session, which can be achieved by the cooking appliance I.

[0044] The visual display of browning levels can vary, but the optimal approach is to display the gradient of brown shades. Each shade represents a different browning level on the foodsurface. The visual display can be shown on the interface of the cooking appliance I, a mobile device, TV, personal computer, or similar. The visual display can be shown as prompted by the user or shown throughout the entire cooking session. The visual display is refreshed as prompted by the user. If shown throughout the entire cooking session, it typically refreshes when the processing unit II sends a new recognition result Vi.

Claims

Claims1. A method for handling a cooking appliance, in particular an oven with which a user can control and affect the browning level of the food to be cooked, characterized in that it comprises the following steps: a) The user inserts the food inside the cooking appliance (I) and selects a cooking program and desired parameters via the respective interface; b) The oven forwards the data from step a) to a processing unit (II), which stores the data in a storage (III), after which the processing unit (II) selects at least one model (IV) and starts a new cooking session at the selected time; c) Optimal lighting with a lighting means (6) and sensing with the device (7) to capture and store images and / or videos of the food to be cooked inside the cooking appliance (I) in regular time intervals, wherein the cooking appliance (I) forwards data to the processing unit (II), which stores it in the storage (III); d) The model (IV) obtains the latest data from the storage (III) and processes it in the pre-processing layer (IV ; e) The data processed in the pre-processing unit (II) is transferred to the recognition layer IV2, which by means of a suitable algorithm prepares the recognition proposal IV3; f) The post-processing unit (V) obtains the previous results of browning recognition from the storage (III) and the recognition proposal (IV3) from the model (IV), after which it dynamically improves the recognition proposal (IV3) from the model (IV) and creates the most accurate recognition result (V ; g) The processing unit (II), based on the recognition result (V , determines the next steps, which can vary based on the following scenarios; h) Once the cooking inside the cooking appliance (I) ends, recognition with the device (7) to capture and store images and / or video of the food to be cooked ends immediately, while the illumination by means of the lighting means (6) stops with a delay, either immediately or gradually by reducing the intensity to zero, while a heat sink (6c) remains operational until the oven cools down.

2. A cooking appliance in particular for carrying out the method according to any of the proceeding claims, characterised in that, in the area above an upper wall (5) of walls (2) of a cooking chamber (1), at least one lighting means (6) is arranged, comprising a light guide rod (6a), to illuminate a tray (4) for accommodating food that is to be thermally treated, and to illuminate the interior of the cooking chamber (1), and at least one device (7) to capture and store images and / or videos, wherein the lighting means (6) and the device (7) to capture and store images and / or videos are arranged in parallel or in series in close vicinity next to each other, and where the light guide rod (6a) comprises a free end (6e) protruding into the cooking chamber (1) said free end (6e) formed with an inclination (11) comprising an angle ((3) in relation to the longitudinal centre line of the lighting means (6), wherein the radiation light flux, which exits through the free end (6e) of the light guide rod (6a), faces the area of the most distant corner in such a manner that the said area is illuminated as much as possible, and wherein the lighting means (6) and the device (7) to capture and store images and / or videos, or the longitudinal body axes of the lighting means (6) and the device (7), are, in relation to the horizontal plane of the cooking appliance, tilted at an angle (a) in the direction from the back wall of the walls (2) toward the door (3), in such a manner that the ends of the light guide rod (6a) and the device (7), which extends into the cooking chamber (2), face a back wall of the walls (2).

3. A cooking appliance according to claim 2, characterised in that the free end (6e) of the light guide rod (6a), on the side diametrically opposite to the inclination (11), further comprises a surface (12) finished in such a manner that it at least partly transmits light along the light guide rod (6a), wherein the surface (12), along the perimeter of the light guide rod (6a), spreads for at least 180° and symmetrically in relation to the surface (12) and wherein the surface (12), in the longitudinal direction of the light guide rod (6a), extends from the tip (11’) of the inclination (11) for a length (I), which corresponds to the longest dimension of the surface (12) according to the longitudinal centre line of the light guide rod (6a).

4. A cooking appliance according to claims 2 and 3, characterised in that the inclination (11) and the surface (12) are formed as a translucent surface, for instance a satinised,matt, or frosted surface.

5. A cooking appliance according to claims 2 to 4, characterised in that the said angle (a) of the inclination of the light guide rod (6a) and the device (7) is identical.

6. A cooking appliance according to claims 2 to 5, characterised in that the said angle (a) of the inclination falls within a range of approximately 0° to approximately 45°.

7. A cooking appliance according to claims 2 to 5, characterised in that the angle (a) of the inclination of the light guide rod (6a) falls within a range of approximately 0° to approximately 45°, while the angle of the inclination of the device (7) differs from the angle (a) of the inclination of the light guide rod (6a).

8. A cooking appliance according to claims 2 to 7, characterised in that the said angle (3) falls within a range of approximately 30° to approximately 60°.

9. A cooking appliance according to claims 2 to 8, characterised in that in and / or on at least one side wall (8, 9) of the walls (2), at least one additional lighting means (10) is provided to illuminate the interior of the cooking chamber (1).

10. A cooking appliance according to claims 2 to 9, characterised in that at least one of the lighting means (6; 10) is selected as a light-emitting diode (LED) and at least one of the lighting means (6; 10) is, for example, selected as a light guide rod (6a).

11. A cooking appliance according to claims 2 to 10, characterised in that the lighting means (6) is arranged on / inside the device (7) to capture and store images and / or videos in such a way that it at least partly surrounds the lens of the device (7).

12. A cooking appliance according to claims 2 to 11, characterised in that the light guide rod (6a) is rotated around its longitudinal body axis in such a way that most of the light that exits from the free end (6e) faces the area of the most distant corner of the cooking chamber (1).

13. A system for sensing the browning level of the food to be cooked, characterized in that it comprises: a) A cooking appliance (I) with a lighting means (6) to illuminate the inside of the oven and a sensor device (7) to capture and store images and / or video of the food to be cooked inside the cooking appliance (I) in regular time intervals; b) A processing unit (II), which is coupled with the cooking appliance (I), wherein the processing unit (II) comprises a storage (III), at least one model (IV), and a postprocessing unit (V); c) Wherein at least one said model (IV) comprises at least one pre-processing layer (IVi) and at least one recognition layer (IV2), wherein the pre-processing layer (IV prepares data for entry into the recognition layer (IV2), which creates a recognition proposal (IV3), which is then directed to the post-processing unit (V) for further analysis and processing; d) Wherein the processing unit (II) obtains the previous result from the storage (III) and the recognition proposal (IV3) from the model (IV) in order to dynamically improve the recognition proposal (IV3) to obtain the most accurate recognition result (Vi.)14. The system according to claim 12, characterized in that at least one or all of the units (II, III, IV, V) can be integrated in the cooking appliance (I) or in a separate external device, for instance a personal computer, mobile device, cloud storage, TV, and similar.

Images