JP7065305B2 - Cooker - Google Patents

Description

The present invention relates to a cooking device that heats a cooking object.

Conventionally, there has been known a cooking cooker in which a cooking object is photographed by a camera and the heating of the cooking object is controlled based on the image of the cooking object captured in the photographed image (see, for example, Patent Document 1).

Japanese Unexamined Patent Publication No. 2015-68542

However, in the case of the cooking device described in Patent Document 1, when the light environment around the cooking device changes, the color tone of the image to be cooked in the image taken by the camera changes. For example, when the curtain is opened during cooking and the light of the setting sun enters the kitchen and the light hits the cooking object, the entire photographed image, that is, the image of the cooking object may become reddish (hue changes). Further, for example, when a light is turned on during cooking, the entire photographed image, that is, the image to be cooked may be whitish (saturation is reduced) due to the light of the light. If the color tone of the cooking object shown in the photographed image changes or differs due to such a change in the light environment, it may not be possible to properly cook the cooking object based on the image of the cooking object reflected in the photographed image.

Therefore, according to the present invention, in a cooking device that controls heating of a cooking object based on an image of the cooking object captured by a camera, the influence of changes in the light environment is suppressed and the cooking object is appropriately heated. That is the issue.

In order to solve the above-mentioned problems, according to one aspect of the present invention,
The container mounting part where the container containing the cooking object is placed, and
A heating unit that heats the container placed on the container mounting unit from below,
A camera that shoots the cooking target,
A cooking target state estimation unit that extracts a feature amount from an image of a cooking target captured in an image taken by the camera at a predetermined timing interval and estimates the state of the cooking target based on the extracted feature amount.
It has a heating control unit that controls the heating unit based on the state of the cooking object estimated by the cooking target state estimation unit.
The cooking target state estimation unit
When the amount of change in the feature amount from the first timing to the second timing immediately after the first timing exceeds a predetermined threshold change amount.
The state of the cooking target at the third timing after the second timing, the feature amount at the first timing, the feature amount at the third timing, and the feature at the second timing. A cooker is provided that is configured to estimate based on the difference between the amount and the amount.

According to the present invention, in a cooking device that controls heating of a cooking object based on an image of the cooking object captured by a camera, the influence of changes in the light environment is suppressed and the cooking object is appropriately heated. be able to.

These aspects and features of the invention become apparent from the following description relating to the preferred embodiments of the accompanying drawings. In this drawing,
Schematic perspective view of the cooking apparatus according to one embodiment of the present invention. The figure which shows the structure of the cooking apparatus roughly. Block diagram showing the control system of the cooking device The figure which shows the change of the image of a cooking object in one example of cooking The figure which shows the change of the image of a cooking object in another example of cooking The figure which shows the change of the image of the cooking object exposed to the ambient light in the cooking shown in FIG. The figure which shows the change of the image of the cooking object exposed to the ambient light in the cooking shown in FIG. The figure which shows the change curve of the feature quantity influenced by the change of a light environment, and the change curve of a feature quantity corrected so that the influence of a change of a light environment is reduced. The figure which shows the change curve of the feature amount affected by the change of the light environment and the change curve of the feature amount corrected to reduce the influence of the change of the light environment when the timing interval is changed by half.

The heating cooker according to one aspect of the present invention includes a container mounting portion on which a container containing a cooking object is placed, a heating portion that heats the container mounted on the container mounting portion from below, and the above. A cooking target that extracts a feature amount from a camera that shoots the cooking target and an image of the cooking target that appears in the image taken by the camera at a predetermined timing interval, and estimates the state of the cooking target based on the extracted feature amount. It has a state estimation unit and a heating control unit that controls the heating unit based on the state of the cooking target estimated by the cooking target state estimation unit, and the cooking target state estimation unit has a first timing. When the amount of change in the feature amount from the first timing to the second timing immediately after the first timing exceeds a predetermined threshold change amount, the state of the cooking target in the third timing after the second timing is determined. It is configured to estimate based on the difference between the feature amount at the first timing, the feature amount at the third timing, and the feature amount at the second timing. ..

According to one aspect of the present invention, in a cooking device that controls heating of a cooking object based on an image of the cooking object captured by a camera, heating of the cooking object is suppressed by suppressing the influence of changes in the light environment. Can be done properly.

For example, the feature amount may be at least one of hue, saturation, and lightness. Since the feature amount is the hue, saturation, and lightness related to the hue of the image of the object to be cooked which is easily affected by the change in the color environment, the change amount of the feature amount from the first timing to the second timing is predetermined. When the threshold change amount is exceeded, the state of the cooking target at the third timing is changed to the feature amount at the first timing, the feature amount at the third timing, and the second timing. It is appropriate to estimate based on the difference from the feature amount in.

For example, at least one of the predetermined timing interval and the predetermined threshold change amount may be changed based on the heating output of the heating unit. Thereby, the state of the cooking object can be estimated with high accuracy.

For example, the cooking device has a temperature sensor that detects the temperature of the container placed on the container mounting portion, and the cooking target state estimation unit has the detection temperature at the first timing and the said. When the detected temperature at the second timing is substantially the same, the state of the cooking target at the third timing is the feature amount at the first timing and the feature at the third timing. It may be configured to estimate based on the difference between the quantity and the feature quantity at the second timing. Thereby, when the amount of change in the feature amount from the first timing to the second timing exceeds a predetermined threshold change amount, it can be guaranteed that the cause is the change in the light environment.

For example, the cooking device has a weight sensor that detects the weight of the container placed on the container mounting portion, and the cooking target state estimation unit has the detected weight at the first timing and the said. When the detected weight at the second timing is substantially the same, the state of the cooking target at the third timing is the feature amount at the first timing and the feature at the third timing. It may be configured to estimate based on the difference between the quantity and the feature quantity at the second timing. Thereby, when the amount of change in the feature amount from the first timing to the second timing exceeds a predetermined threshold change amount, it can be guaranteed that the cause is the change in the light environment.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

FIG. 1 is a schematic perspective view of a cooking device according to an embodiment of the present invention. Further, FIG. 2 is a diagram schematically showing the configuration of the cooking device. Furthermore, FIG. 3 is a block diagram showing a control system of the cooking device. In the figure, the X-axis direction indicates the width direction of the cooking cooker, the Y-axis direction indicates the depth direction, and the Z-axis direction indicates the height direction.

As shown in FIG. 1, the cooking cooker 10 has a main body 12, and as an upper portion of the main body 12, a top plate 14 on which a container C containing a cooking object T such as a stew is placed is placed. Have.

Further, in the case of the present embodiment, the heating cooker 10 is an induction heating cooker, and is placed on a container mounting portion (container mounting area) in the top plate 14 as shown in FIGS. 1 and 2. The heating coils 16A to 16C are arranged below the container C, that is, in the main body 12, as a heating portion of the cooking device 10. The heating coils 16A to 16C induce and heat the container C placed on the top plate 14 from below.

As an operation unit for the user to operate each of the heating coils 16A to 16C, a plurality of touch keys 18A to 18C are provided on the portion of the top plate 14 on the front surface 10a side of the cooking device 10. For example, the touch key 18A starts or stops the heating of the corresponding heating coil 16A. Also, the heating level (eg, 4 steps) is adjusted.

Further, a plurality of output display units 20A to 20C for displaying the output states (heating levels) of the heating coils 16A to 16C are also provided on the portion of the top plate 14 on the front surface 10a side of the cooking device 10.

In addition, in order to set the heating by the heating coils 16A to 16C in detail, the setting unit 22 is provided on the front surface 10a side of the cooking device 10. The setting unit 22 can be taken in and out of the main body 12, and sets the setting key 22a for setting the heating by the heating coils 16A to 16C in detail, the setting contents, the detailed state of the heating coils 16A to 16C, and the like. A setting display unit 22b for displaying is provided. The setting unit 22 sets the heating temperature, heating time, timer, and the like of the heating coils 16A to 16C.

Further, the cooking cooker 10 is provided with a notification unit 24 such as a speaker for notifying the user on the front surface 10a side of the cooking cooker 10.

Further, the cooking cooker 10 has a plurality of temperature sensors 26A to 26C for detecting the temperature of the container C placed on the top plate 14, and a weight for detecting the weight of the container C placed on the top plate 14. It has a sensor 28.

In the case of the present embodiment, the plurality of temperature sensors 26A to 26C are arranged below the top plate 14 and at the center of each of the heating coils 16A to 16C. That is, for example, the temperature sensor 26A detects the temperature on the bottom side of the container C, which is arranged above the heating coil 16A and is induced and heated by the heating coil 16A, via the top plate 14.

In the case of the present embodiment, the weight sensor 28 uses the weight of the object placed on the top plate 14, that is, the weight of the plurality of containers C when there are a plurality of containers C, and the cooking contained in each container C. Detect the sum with the weight of the object.

Further, the cooking device 10 has a camera 30 for photographing the cooking target T on the top plate 14 from above.

The camera 30 is a camera capable of photographing, for example, a still image, a moving image, an infrared image, and the like, and photographs the entire top surface 14a of the top plate 14. The camera 30 is attached to a ceiling or a range hood (not shown) located above the cooking device 10. In the case of this embodiment, the camera 30 is attached to the range hood via a magnet.

In the case of the present embodiment, the camera 30 is configured to wirelessly communicate with the control unit 50 of the cooking cooker 10. Therefore, as shown in FIG. 2, the camera 30 is provided with the antenna 32, and the antenna 34 is provided in the main body 12 of the cooking cooker 10. In the case of the present embodiment, the camera 30 and the control unit 50 are connected to each other via the router device 36. That is, the camera 30 and the control unit 50 are wirelessly connected to the local area network 38 centered on the router device 36. Further, the router device 36 is connected to the Internet 40. Further, the router device 36 is connected to the user's mobile terminal 42 so as to be capable of wireless communication.

From here on, the operation of the cooking cooker 10, that is, the control performed by the control unit 50 will be described with reference to FIG.

The control unit 50 of the cooking device 10 is a processing device such as a CPU, and performs various functions by executing a program stored in the storage unit 60 such as a ROM, RAM, and a hard disk (for example). , To function as an image processing unit described later). As shown in FIG. 3, the control unit 50 is configured to receive signals corresponding to the user's operation on the touch keys 18A to 18C and the setting key 22a of the setting unit 22 from these. The control unit 50 is also configured to receive signals corresponding to the temperatures detected by the temperature sensors 26A to 26C, respectively. The control unit 50 is further configured to receive a signal corresponding to the weight detected by the weight sensor 28. The control unit 50 is configured to receive a captured image (data) captured by the camera 30 from the camera 30.

In addition, the control unit 50 of the cooking cooker 10 is configured to control the heating coils 16A to 16C, the output display units 20A to 20C, the setting display unit 22b, and the notification unit 24.

For example, the control unit 50 controls the output of the heating coils 16A to 16C so that the temperature sensors 26A to 26C continue to detect the heating temperature set by the user via the setting key 22a of the setting unit 22. Thereby, for example, when frying, the oil contained in the container C can be maintained at a constant temperature.

Further, for example, when water is contained in the container C (when the user sets the automatic boiling mode via the setting unit 22), the control unit 50 heats when the temperature sensors 26A to 26C detect the boiling temperature. The coils 16A to 16C are stopped, and the user is notified via the notification unit 24 that the boiling point has been completed.

Further, the control unit 50 is configured to perform various controls based on the captured image of the camera 30, specifically, the image of the cooking target T captured in the captured image.

Therefore, the control unit 50 of the cooking cooker 10 includes an image acquisition unit 52 that acquires a captured image from the camera 30, an image processing unit 54 that processes (corrects) the acquired image, and the corrected captured image. The cooking target state estimation unit 56 that estimates the state of the cooking target T based on the image of the cooking target T in the imaged container C, and the heating as a heating control unit that controls the heating coils 16A to 16C based on the estimation result. It has a coil control unit 58 (that is, the control unit 50 functions as an image acquisition unit 52, an image processing unit 54, a cooking target state estimation unit 56, and a heating coil control unit 58).

The image acquisition unit 52 of the control unit 50 of the cooking device 10 captures an image taken by the camera 30, for example, a top plate 14 in which a container C for accommodating a cooking target T being cooked is placed. Get an image (data). In the case of this embodiment, the camera 30 takes pictures at predetermined timing intervals.

The image processing unit 54 performs image processing on the captured image of the camera 30 acquired by the image acquisition unit 52. For example, geometric correction such as keystone correction and rotation correction is performed.

The cooking target state estimation unit 56 is configured to estimate the state of the cooking target based on the feature amount of the image of the cooking target reflected in the captured image. The estimation method will be specifically described.

The image of the cooking object shown in the image taken by the camera 30 is an image of the cooking object seen from above. A feature amount is extracted from the image of the cooking target viewed from above at a predetermined timing interval (in the case of the present embodiment, the same interval as the shooting interval of the camera), and the cooking target is based on the extracted feature amount. The state estimation unit 56 estimates the state of the cooking target.

Here, the "feature amount" refers to a numerical value of the characteristic portion of the image of the cooking object that changes with the change of the state of the cooking object. For example, when the hue of the image of the cooking object changes with the change of the state of the cooking object, the "feature amount" is hue, saturation, lightness, or the like.

By giving an example, the estimation of the state of the cooking object based on the feature amount of the image of the cooking object captured in the image taken by the camera 30 will be described.

FIG. 4 shows a change in the image of the object to be cooked in the photographed image in one example of cooking. Specifically, the image of the cooking target CT1 in the photographed images taken at different timings T1 to T5 is shown.

The cooking target CT1 shown in FIG. 4 is a fried egg. With the passage of heating time, the image of the white part CTa changes from transparent to white (transparency is reduced). In FIG. 4, the inner bottom surface of the container C1 (frying pan) is shown by cross-hatching. Further, the photographed image P5 taken at the timing T5 shows an image of a fried egg in which the entire white meat portion CTa is completely discolored to white, that is, an image of a fried egg when cooking is completed.

That is, as shown in FIG. 4, as the feature amount of the cooking target CT1, the saturation and the lightness of the white color of the white meat portion CTa in the fried egg increase with the passage of the heating time. Since the feature amount changes with the passage of the heating time, the cooking target state estimation unit 56 can extract the feature amount and estimate the state of the cooking target CT1 which is a fried egg based on the extracted feature amount.

FIG. 5 shows a change in the image of the object to be cooked in the photographed image in another example of cooking.

The cooking target CT2 shown in FIG. 5 is a stew. With the passage of heating time, the image of bubbles generated on the liquid surface increases in number and size. Therefore, as a feature amount of the CT2 to be cooked, the saturation of the color (cream color) on the entire liquid surface of the stew decreases with the passage of the heating time. Since the feature amount changes with the passage of the heating time, the cooking target state estimation unit 56 can extract the feature amount and estimate the state of the cooking target CT2 which is a stew based on the extracted feature amount.

The feature amount extracted by the cooking target state estimation unit 56 when estimating the state of the cooking target is preferably a feature amount that greatly changes with a change in the heating time. Therefore, the feature amount used for estimating the state may be predetermined for each cooking object. For example, in the storage unit 60, the type of the cooking target and the feature amount to be extracted are stored in association with each other. The type of cooking object is taught by the user. For example, a list of types of cooking targets is displayed on the setting display unit 22b, and the user selects the type of cooking target to be cooked from the list via the setting key 22a, so that the type of cooking target is set to the user. Is taught by. Based on the type of the cooking target taught by the user, the cooking target state estimation unit 56 extracts the feature amount corresponding to the type, and estimates the state of the cooking target based on the extracted feature amount.

Further, the state of the cooking target may be estimated based on a plurality of features instead of one feature. For example, as the two features, hue and saturation, lightness and size, saturation and shape, and the like may be used.

Further, the state of the cooking object may be estimated based on the change in the feature amount of the image of the cooking object in the photographed image. For example, even if the state of the cooking target at a certain timing is estimated based on a plurality of features of the image of the cooking target appearing in each of the captured images taken at the timing before that, that is, based on the change in the features. good. For example, when the feature amount is the size or shape of the image to be cooked, the state may be estimated based on the change rate of the size or shape. Since the change in the feature amount (based on a plurality of feature amounts) is estimated and the state of the cooking target can be estimated, highly reliable estimation can be performed.

Furthermore, the state of the cooking object may be estimated based on the feature amount of the image of the cooking object in the captured image and based on at least one of the detected temperature of the temperature sensors 26A to 26C and the detected weight of the weight sensor 28. .. For example, as the heating time elapses, the temperature of the container, that is, the temperature of the object to be cooked contained in the container rises. By detecting the temperature rise by the temperature sensors 26A to 26C and considering the detected temperature, it is possible to guarantee the certainty in estimating the state of the cooking object based on the feature amount. Further, for example, when the object to be cooked is a liquid, the weight of the object to be cooked decreases due to evaporation with the passage of heating time. By detecting the weight loss by the weight sensor 28 and considering the detected weight, it is possible to guarantee the certainty in estimating the state of the cooking object based on the feature amount.

In this way, the cooking target state estimation unit 56 estimates the state of the cooking target based on the feature amount of the image of the cooking target captured by the camera 30. Therefore, it is easily affected by changes in the light environment around the cooking device 10. In particular, when the feature amount is the hue, saturation, or lightness related to the hue of the image to be cooked, it is greatly affected.

Specifically, when the light environment around the cooking device 10 changes, the color tone of the image to be cooked in the image taken by the camera 30 changes. For example, if the curtain is opened during cooking and the light of the setting sun enters the kitchen, the light may cause the entire captured image, that is, the image to be cooked, to become reddish (change in hue). Further, for example, when a light is turned on during cooking, the image to be cooked may become whitish (saturation is reduced) due to the light of the light.

For example, FIG. 6 shows a change in the image of a cooking object exposed to ambient light in the cooking shown in FIG.

As shown in FIG. 6, the captured images P12 to P15 captured after the timing T2 are affected by the change in the light environment around the cooking cooker 10 between the timing T1 and the timing T2. For example, a curtain is opened between the timing T1 and the timing T2, whereby sunlight (turbulent light) hits the cooking object.

As a result, when FIG. 4 and FIG. 6 are compared, the image of the white meat portion CTa of the cooking target CT1 exposed to the ambient light reflected in the captured images P12 to P14 shown in FIG. 6 at the timings T2 to T4 is shown in FIG. Compared with the image of the white meat portion CTa of the cooking target CT1 not exposed to the ambient light shown in the photographed images P2 to P4, the image is whiter (low transparency). Therefore, when the feature amount is saturation for white, the feature amount extracted from the image of the cooking object exposed to the ambient light is larger than the feature amount extracted from the image of the cooking object not exposed to the disturbance light. are doing.

For example, the image of the white meat portion CTa of the cooking target CT1 exposed to the disturbance light reflected in the captured image P14 at the timing T4 shown in FIG. 6 is the cooking not exposed to the disturbance light reflected in the captured image P5 at the same timing T4 shown in FIG. Compared to the image of the white part CTa of the target CT1, the saturation and lightness of the white color, which is a feature amount, is high (the transparency is low). Based on such a feature amount, it can be presumed that at the timing T4, the cooking object is sufficiently heated when it is exposed to ambient light, even though it is actually in a state of being burnt.

Further, for example, FIG. 7 shows a change in the image of a cooking object exposed to ambient light in the cooking shown in FIG.

As shown in FIG. 7, the captured images P17 to P20 captured after the timing T2 are affected by the change in the light environment around the cooking cooker 10 between the timings T1 and T2. For example, a curtain is opened between the timing T1 and the timing T2, whereby sunlight (turbulent light) hits the cooking object.

As a result, when FIG. 5 and FIG. 7 are compared, the image of the cooking target CT2 exposed to the ambient light reflected in the captured images P18 to P20 shown in FIG. 7 at each of the timings T3 to T5 is the captured image P8 shown in FIG. Compared to the image of the cooking target CT2 that is not exposed to the ambient light reflected in P10, the image of bubbles is less, and therefore the image is bright cream color (so-called white tobi is generated). Therefore, when the feature amount is the saturation for cream color, the feature amount extracted from the image of the cooking object exposed to the ambient light is compared with the feature amount extracted from the image of the cooking object not exposed to the disturbance light. Is increasing.

For example, the image of the cooking target CT2 exposed to the disturbance light reflected in the captured image P18 at the timing T3 shown in FIG. 7 is an image of the cooking target CT2 not exposed to the disturbance light reflected in the captured image P8 at the same timing T3 shown in FIG. Compared to, the color saturation of cream, which is a feature, is high (because there is no image of bubbles). Based on such features, it can be presumed that at timing T3, despite the weak boiling state in which fine bubbles are actually generated, it is in an unboiling state when exposed to ambient light. ..

When the feature amount changes due to the change in the light environment in this way, the accuracy of estimating the state of the cooking object based on the feature amount decreases. Therefore, when the feature amount changes due to a change in the light environment, the state of the cooking target is estimated so that the state of the cooking target is estimated by the feature amount (corrected feature amount) corrected so as to reduce the influence of the change in the light environment. The unit 56 is configured. This will be specifically described with reference to FIG.

FIG. 8 shows a change curve of the feature amount (two-dot chain line) affected by the change of the light environment and a change curve of the feature amount (corrected feature amount) corrected so as to reduce the influence of the change of the light environment (corrected feature amount). It is a figure which shows (solid line). As a reference, FIG. 8 also shows a change curve (dashed line) of the feature amount when the light environment does not change.

The change curve (dashed-dotted line) shown in FIG. 8 is a feature amount extracted from an image of a cooking object that is affected by a change in the light environment and is reflected in an image taken by the camera 30, which changes with the passage of the heating time t. S (that is, the measured value) is shown. Further, the change curve (solid line) is used for estimating the feature amount SA (corrected feature amount) (that is, the state of the cooking object) corrected so as to reduce the influence of the change in the light environment, which changes with the passage of the heating time t. The value to be) is shown. The change curve (dashed-dotted line) shows the feature amount SO extracted from the image of the cooking target when the light environment does not change, which changes with the passage of the heating time t.

Further, FIG. 8 shows the feature amount S when the light environment changes between the timing T1 and the timing T2, for example, when the cooking object continues to be exposed to the ambient light from the timing between the timing T1 and the timing T2. The change curves of SA and SO are shown.

First, the cooking target state estimation unit 56 determines that when the amount of change in the feature amount between continuous timings exceeds a predetermined threshold change amount, the light environment has changed during this timing. For example, in the case of the example shown in FIG. 8, the difference ΔSd (absolute value) between the feature amount S1 at the timing T1 and the feature amount S2 at the timing T2 is a predetermined value obtained experimentally or theoretically in advance. The threshold change amount ΔSth is exceeded. That is, the amount of change in the feature amount from the timing T1 to the timing T2 exceeds the predetermined threshold change amount ΔSth. Therefore, after the timing T2, the cooking target state estimation unit 56 treats the feature quantities S2 to S5 extracted from the image of the cooking target as the feature quantities affected by the change in the light environment.

As shown in FIG. 8, when the light environment changes between the timing T1 and the timing T2 (when the cooking object is exposed to ambient light), the feature quantities S2 to S5 extracted from the images of the cooking objects at the timings T2 to T5. Is uniformly increased by the increment ΔSL as compared with the feature quantities SO2 to SO5 extracted when the light environment has not changed (the cooking target is not exposed to ambient light). However, this increment ΔSL cannot be calculated because the feature quantities SO2 to SO5 are not the feature quantities that can be actually extracted (because they cannot be actually measured).

Therefore, the cooking target state estimation unit 56 sets the cooking target states at the timings T2 to T5 after the timing T2 at the feature quantities S1 at the timing T1 and the feature quantities S2 to S5 and the timing T2 at the timings T2 to T5. It is configured to estimate based on the difference ΔS2 to ΔS5 with the feature amount S2. Specifically, in the timings T2 to T5 after the timing T2, the corrected feature amounts SA2 to SA5, which are the correction values of the extracted feature amounts S2 to S5 and are the sum of the feature amounts S1 and the differences ΔS2 to ΔS5. Estimate the state using.

For example, the corrected feature amount SA2 used for estimating the state of the cooking target at the timing T2 is S1 + ΔS2. Since the difference ΔS2 is zero (because S2-S2 = 0), the correction feature amount SA2 is S1. Further, the correction feature amount SA3 used for the estimation at the timing T3 is S1 + ΔS3. The difference ΔS3 is S3-S2. Further, the correction feature amount SA4 used for the estimation at the timing T4 is S1 + ΔS4. The difference ΔS4 is S4-S2. The correction feature amount SA4 used for the estimation at the timing T5 is S1 + ΔS5. The difference ΔS5 is S5-S2.

That is, the timing Tn (after the timing T2) used when the difference ΔSd (absolute value) between the feature amount S1 at the timing T1 and the feature amount S2 at the timing T2 exceeds the predetermined threshold change amount ΔSth. The corrected feature amount SAN used for estimating the state of the cooking target in (n is an integer of 2 or more) can be expressed by Equation 1. Sn is a feature amount at timing Tn.

The corrected feature amount SA calculated in this way is extracted from the image of the cooking target with respect to the feature amount SO when the light environment has not changed (the cooking target is not exposed to ambient light), and the light environment is calculated. It is possible to take a value closer to that of the feature amount S affected by the change. Therefore, the corrected feature amount SA is less affected by changes in the light environment than the feature amount S (actually measured value) extracted from the cooking target. Therefore, if such a corrected feature amount SA is used, the state of the cooking target can be estimated with high accuracy (compared to the case where the feature amount S is used).

The difference (error) between the corrected feature amount SA and the feature amount SO when the light environment has not changed (the cooking target is not exposed to ambient light) is the time (timing interval) Tp between timings. The shorter it is, the smaller it can be. For example, FIG. 9 shows a change curve of the feature amount S affected by a change in the light environment when the time between timings is changed by half, and a feature corrected so as to reduce the influence of the change in the light environment. The change curve of the quantity SA is shown.

Comparing FIGS. 8 and 9, when the timing interval Tp is halved, there is a difference between the corrected feature amount SA and the feature amount SO when the light environment has not changed (the cooking target is not exposed to ambient light). It can be seen that the difference (error) becomes small. That is, the smaller the timing interval Tp, the more the influence of the change in the light environment can be reduced, and as a result, the state of the cooking object can be estimated with higher accuracy.

In the example shown in FIG. 8, the cooking target state estimation unit 56 determines that the difference ΔSd between the feature amount S1 at the timing T1 and the feature amount S2 at the timing T2 exceeds the predetermined threshold change amount ΔSth. , It is judged that the light environment changed during these timings. However, there is a slight possibility that the amount of change in the feature amount between continuous timings will exceed a predetermined amount of change without changing the light environment.

For example, when the cooking target is a liquid such as a stew, when the user stirs the cooking target in a weakly boiling state, the generation of bubbles on the liquid surface is temporarily weakened. Therefore, the difference between the feature amount at the timing before the user stirs and the feature amount at the timing after stirring may exceed a predetermined threshold change amount. In addition, the appearance of the object to be cooked changes significantly when seasonings and ingredients are added. The difference between the feature amount at the timing before the addition and the feature amount at the timing after the addition may exceed a predetermined threshold change amount. That is, when some change is given by the user to the cooking target during cooking, the amount of change in the feature amount between continuous timings may exceed a predetermined threshold change amount.

Therefore, when the amount of change in the feature amount between continuous timings exceeds a predetermined amount of change, it is preferable to guarantee that the cause is the change in the optical environment.

Therefore, the temperature of the cooking target (container) may be detected by using the temperature sensors 26A to 26C at each timing for estimating the state of the cooking target. When the amount of change in the feature amount between consecutive timings exceeds a predetermined threshold change amount and the cause is a change in the optical environment, the amount of change in temperature between these timings is practically zero. .. For example, in the case of the example shown in FIG. 8, when the difference ΔSd between the feature amount S1 at the timing T1 and the feature amount S2 at the timing T2 exceeds a predetermined threshold change amount ΔSth, the cause is the light environment. In the case of a change, the temperature detected by the temperature sensor at the timing T1 and the temperature detected at the timing T2 are substantially the same. This is because the temperature of the object to be cooked does not substantially change due to changes in the light environment.

That is, when the amount of change in the feature amount between consecutive timings exceeds a predetermined threshold change amount, if the amount of change in temperature between the timings is substantially zero, the light environment changes between the timings. It can be judged that it has changed. On the other hand, when the amount of temperature change between the timings is not substantially zero (for example, when the amount of temperature change exceeds a predetermined threshold temperature change amount), some change is made by the user with respect to the cooking target. It can be judged that it was given.

Therefore, in the cooking target state estimation unit 56, the amount of change in the feature amount between consecutive timings exceeds a predetermined threshold change amount, and in addition, the amount of change in the temperature of the cooking target between the timings is substantially. If it is zero, it is judged that the light environment has changed between continuous timings. This makes it possible to appropriately estimate the state of the cooking target.

Instead of or in addition to using the detection temperature of the temperature sensor, the weight sensor 28 may detect the weight of the cooking target (container) at each timing of estimating the state of the cooking target. Similar to the temperature of the object to be cooked, the weight of the object to be cooked does not substantially change due to changes in the light environment. For example, in the case of the example shown in FIG. 8, when the difference ΔSd between the feature amount S1 at the timing T1 and the feature amount S2 at the timing T2 exceeds a predetermined threshold change amount ΔSth, the cause is the light environment. In the case of a change, the detected weight of the weight sensor at the timing T1 and the detected weight at the timing T2 are substantially the same.

That is, as with the temperature of the object to be cooked, when the amount of change in the feature amount between consecutive timings exceeds a predetermined threshold change amount, the amount of change in weight between the timings is substantially zero. , It can be judged that the light environment has changed between the timings. Therefore, in the cooking target state estimation unit 56, the amount of change in the feature amount between consecutive timings exceeds a predetermined threshold change amount, and in addition, the amount of change in the weight of the cooking target between the timings is substantially. If it is zero, it is judged that the light environment has changed between continuous timings. This makes it possible to appropriately estimate the state of the cooking target.

Returning to FIG. 3, the heating coil control unit 58 controls the outputs of the heating coils 16A to 16C based on the state of the cooking target estimated by the cooking target state estimation unit 56. For example, when the cooking target is estimated to be evaporating by the cooking target state estimation unit 56, the heating coil control unit 58 lowers the output level of the heating coil or stops the heating coil.

As described above, according to the present embodiment as described above, in the cooking device 10 that controls the heating of the cooking object based on the image of the cooking object captured by the camera 30, the influence of the change in the light environment is suppressed. It is possible to properly heat the cooking object.

Although the present invention has been described above with reference to the above-described embodiments, the present invention is not limited to the above-mentioned embodiments.

For example, in order to estimate the state of the cooking target with high accuracy, a predetermined timing interval Tp at a plurality of timings for estimating the state may be changed based on the heating output of the heating coil. The higher the heating output (heating level) of the heating coil that heats the object to be cooked, the more rapidly the state of the object to be cooked changes. Therefore, it is preferable that the higher the heating output of the heating coil, the shorter the predetermined timing interval Tp, and the more frequently the state of the cooking target is estimated.

In addition to or instead of changing the predetermined timing interval Tp, the predetermined threshold change amount ΔSth to be compared with the change amount of the feature amount between consecutive timings is changed based on the heating output of the heating coil. You may. The higher the heating output (heating level) of the heating coil, the more accurate it is necessary to estimate the state of the cooking object based on the feature amount of the image of the cooking object in the photographed image. For that purpose, a feature amount (corrected feature amount) in which the influence of the light environment is further reduced is required. Otherwise, if the estimation accuracy of the state of the cooking object becomes low, the heating output control of the heating coil based on the estimation result cannot be appropriately performed, and the cooking object may be excessively heated or insufficiently heated. Therefore, the higher the heating output of the heating coil, the smaller the influence of the change in the light environment on the feature amount, but in order to reduce the influence, the threshold change amount ΔSth for determining the occurrence of the change in the light environment. It is preferable to make it smaller.

The predetermined timing interval Tp and the predetermined threshold change amount ΔSth may be different for each type of cooking object. In this case, a predetermined timing interval Tp and a predetermined threshold change amount ΔSth are obtained in advance experimentally or theoretically for each cooking object. Then, the type of the cooking target, the predetermined timing interval Tp, and the predetermined threshold change amount ΔSth are stored in association with each other in the storage unit 60. When the type of the cooking target is taught by the user as described above, the corresponding predetermined timing interval Tp and the predetermined threshold change amount ΔSth are used for estimating the state of the cooking target.

Further, in the case of the above-described embodiment, the heating cooker is an induction heating cooker that induces and heats the container by using a heating coil, but the embodiment of the present invention is not limited to this. For example, the cooking cooker according to the embodiment of the present invention may be a gas cooker. In the case of a gas cooker, the container is placed on the trivet as a container mounting part, and is heated from below by a gas burner as a heating part.

As described above, an embodiment has been described as an example of the technique in the present disclosure. To that end, the accompanying drawings and detailed explanations have been provided. Therefore, among the components described in the attached drawings and the detailed description, not only the components essential for solving the problem but also the components not essential for solving the problem in order to illustrate the technique. Can also be included. Therefore, the fact that those non-essential components are described in the accompanying drawings or detailed description should not immediately determine that those non-essential components are essential.

Further, since the above-described embodiment is for exemplifying the technique in the present disclosure, various changes, replacements, additions, omissions, etc. can be made within the scope of claims or the equivalent thereof.

The specification, drawings, and claims of Japanese Patent Application No. 2017-183986 filed on September 25, 2017 are incorporated herein by reference in their entirety. ..

The present invention is not limited to the induction cooking device, and is applicable to any cooking device that heats a container containing a cooking object.

Claims (5)

A container mounting part on which a container containing a cooking object is placed,
A heating unit that heats the container placed on the container mounting unit from below,
A camera that shoots the cooking target,
A cooking target state estimation unit that extracts a feature amount from an image of a cooking target captured in an image taken by the camera at a predetermined timing interval and estimates the state of the cooking target based on the extracted feature amount.
It has a heating control unit that controls the heating unit based on the state of the cooking object estimated by the cooking target state estimation unit.
The cooking target state estimation unit
When the amount of change in the feature amount from the first timing to the second timing immediately after the first timing exceeds a predetermined threshold change amount.
The state of the cooking target at the third timing after the second timing, the feature amount at the first timing, the feature amount at the third timing, and the feature at the second timing. A cooker that is configured to estimate based on the difference between the amount and. The cooker according to claim 1, wherein the feature amount is at least one of hue, saturation, and lightness. The cooker according to claim 1 or 2, wherein at least one of the predetermined timing interval and the predetermined threshold change amount is changed based on the heating output of the heating unit. It has a temperature sensor that detects the temperature of the container placed on the container mounting portion.
The cooking target state estimation unit
When the detection temperature at the first timing and the detection temperature at the second timing are substantially the same,
The state of the cooking target at the third timing is the difference between the feature amount at the first timing, the feature amount at the third timing, and the feature amount at the second timing. The cooker according to any one of claims 1 to 3, which is configured to be estimated based on the above. It has a weight sensor that detects the weight of the container placed on the container mounting portion.
The cooking target state estimation unit
When the detected weight at the first timing and the detected weight at the second timing are substantially the same,
The state of the cooking target at the third timing is the difference between the feature amount at the first timing, the feature amount at the third timing, and the feature amount at the second timing. The cooker according to any one of claims 1 to 4, which is configured to be estimated based on the above.

Images