DESC:This patent describes the subject matter for patenting with specificity to meet statutory requirements. However, the description itself is not intended to limit the scope of this patent. The principles described herein may be embodied in many different forms.
Illustrative embodiments of the invention now will be described more fully hereinafter with reference to the accompanying drawings, in which some, but not all embodiments of the invention are shown. Indeed, the invention may be embodied in many different forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure will satisfy applicable legal requirements. Like numbers refer to like elements throughout.

The term “communicable coupling” refers to a coupling between devices for enabling exchange or transmission and reception of digital and/or electrical signals between the devices

To solve the problems existing in the conventionally known systems that involve some level of human intervention for controlling a cooking process, the present invention envisions a systemthat monitors the cooking process, such as boiling of a liquid in a vessel being heated,using visual means and utilizes a machine learning module for monitoring the cooking process in a manner that prevents happening of any adverse event that may spoil the food being cooked or pose a danger to a user in vicinity of the cooking vessel. The present invention also envisages that the machine learning module, upon detecting that an adverse event is likely to happen, may control the heating process while continually monitoring the cooking process to eventually stop the process upon detecting that the cooking process is nearing completion. In an embodiment of the present invention, the system can automatically cut off the heat to the cooking vessel by cutting off the source of heat, which could be a gas supply or electricalinput to a heating system used for heating the cooking vessel, thus preventing the storage vessel from heating further.

For the purpose of explanation, the present invention is being explained with reference to a system 100 to control cooking process, as shown in Fig. 1 as well as a method 200, shown in Fig. 3, implemented in the said system. It may, however, be noted that the present invention should not be construed as limited to the described example of cooking process only. The present invention could be used for controlling any type of cooking process such as cooking process using a pressure cooker or baking a product and the like. As shown in Fig. 1, a system 100 is provided to monitor and control a cooking process being carried out using a cooking vessel 110 placed in a heat source 115. Further, as shown in Fig. 2, the system 100 also includes a detector module 120 having an imaging module 125 positioned proximal to the heat source 115 and cooking vessel 110. The imaging module 125 continually monitors the cooking process being carried out in the cooking vessel 110 in a manner that the contents within the cooking vessel 110 is imaged periodicallyso as to help track the progress of the cooking operation, as will be explained in the ensuing description. The imaging module 125 may be a digital camera, such as a standalone single wireless camera, a phone/tablet camera or a DSLR or a mirrorless camera. The present invention envisages that the imaging module 125 captures images of the food product being cooked at pre-determined intervals, at step 202 of the method 200. The pre-determined interval may vary depending on the type of food being cooked. For instance, the interval between capturing of images for a cooking process of boiling milk may be different from that for cooking rice in a vessel. The pre-determined interval may be set by the user or pre-fed in the system based on historical data pertaining to different types of cooking processes. Also, the present invention envisages that the pre-determined interval between capture of two images may vary during the cooking process as well. For instance, in a cooking process involving boiling of milk, the interval may be larger in the beginning but the interval may be reduced once the temperature of the milk is higher as there is high likelihood of milk reaching the boiling temperature faster. For example, for the boiling of milk, during initial heating when the milk is at room temperature or cold, the pre-determined interval may be set to one minute and the pre-determined interval may be reduced to 15 seconds after the heating has been carried out for 3 minutes as the milk may have reached a temperature where it is close to boiling point. Further, the imaging module 125 is communicably coupled to a processor 130, a part of the detector module 120, that includes a machine learning module 135 (Fig. 2) configured to observe each of the clicked images of the cooking process captured by the imaging module 125 and evaluatethe same for ascertaining whether any adverse event is about to happen. The machine learning module 135 is further configured to prevent the adverse event from happening by taking corrective action in the cooking process. The processor 130 may be a general processor or a microcontroller having sufficient processing power to perform machine learning based operations. In addition, the processor is communicably coupled to a communication module 140a that is adapted to be wirelessly connected to a complementary controller unit 140b, which in turn iscoupled to a power source of the heat source 115. The communication module 140a transmits control signals to the controller unit 140b for controlling the operation of the power source of the heat source 115 when the machine learning module 135 determines that there is a likelihood of happening of the adverse event. In particular, the heat source 115 may be switched off, switched on or regulated to control the heat being provided to the cooking vessel 110by transmitting an appropriate control signal as a corrective action in the cooking process, thereby controlling the cooking process. Thus, there may be different control signals for the heat source to be switched off, switched on or regulated to control the heat output of the heat source. In an embodiment of the present invention, the heart source 115 may be a gas stove. In another embodiment of the present invention, the heat source may be an induction heater. The control of the heat source 115 could be carried out with the help of relays that may be used to operate a gas valve in case the heat source is a gas stove. In another embodiment, the relays could control the electrical connection to the induction heater type of heat source 115.

The machine learning module 135evaluatesthe images of the contents of the cooking vessel, received at the pre-determined intervals, based on the training dataset, at step 204 of the method 200,and decides whether the heat source needs to be switched off or on or regulated to control the heat based on the cooking stage of the food being cooked. The decision making is based on the determination of whether an adverse event has happened or about to happen in the cooking process. An “adverse event” with reference to the present invention refers to an unwanted outcome/result of the cooking process. Specifically, the happening of the adverse event may result the cooked food to be in a state that is undesirable. For instance, the adverse event in the case of boiling of a liquid may be the spilling over of the liquid after boiling. In another example, the adverse event in respect of a cooking process being carried out using a pressure cooker would be the food being cooked despite the requisite number of pressure release whistles by the pressure cooker, thereby resulting in the food item being overcooked. Particularly, the adverse event in such a situation may be defined as the number of pressure release operations exceeding the specified or optimal number of pressure release operations required to cook the food ordinarily. In a similar manner, different types of food items to be cooked may have specific adverse events. Further, the present invention envisages that the machine learning module 135 is trained with relevant data (in the form of images and/or videos) of the stages of cooking and adverse events specific to each food item that could be cooked in the cooking vessel 110 so as to provide a reference dataset to the machine learning module 135 to determine the stage at which the food item is. For instance, for a cooking process that involves boiling a liquid, the stages could be a warming stage, a pre-boiling or simmering stage, a boiled/overflow state (adverse stage), post-boiling stage and the like. Similarly, for other types of food items, the stages could be raw stage, medium cooked stage or overcooked stage (adverse stage). For each type of food items, the different stages through which the food item passes during the cooking process as well as adverse stage may be defined and relevant images/videos thereof may be used to train the machine learning module 135. Based on the determination of whether the adverse event may happen, the machine learning module 135 determines, at step 206 of method 200, that a control signal needs to be transmitted to the controller unit to switch off the heat source 115.

The entire process of control of cooking process, as envisaged by the present invention, would now be explained with reference to a cooking process of boiling of milk in the cooking vessel 110, as shown in Fig. 1. In the described embodiment, the machine learning module 135 has been trained with multiple (at least a 100) image and video samples of variations of the adverse events that need to be detected. The machine learning module 135 may be trained using any known machine learning algorithm like, but not limited to, random forest, decision tree, support vector machine or any known deep learning algorithm like but not limited to, convolutional neural networks (CNN), long short term memory networks (LSTMs), Recurrent neural networks (RNNs), Generative Adversarial Networks (GANs), restricted Boltzmann machines (RBMs). For example, for detecting the spilling of a liquid, the machine learning module would be trained using standard machine learning or deep learning algorithms with variations of the liquid spilling/about to spill of the following types:
1. Type of liquid
2. Type of container
3. Level of liquid in the container
4. Imaging of the liquid before and after spilling

Once the module is trained using the image and video samples, it is then tested for accuracy using another set of images and videos covering the scenarios mentioned above. Based on this testing process an accuracy is estimated of the machine learning module. If the accuracy is satisfactory then the module is ready to be deployed. If not, then the training process would have to be repeated with the following potential changes:
1. Number of images and video samples for training to be increased/changed
2. Change the algorithm used for training. In particular, a different Machine learning/deep learning algorithm may be deployed than the one that was originally used and which did not give satisfactory results.

Using this iterative process, the machine learning module is trained and deployed. In use, the detector module 120 continually monitors the cooking vessel having milk being boiled on the heat source 115 by way of clicking images of the milk using the imaging module 125, during the course of boiling process. The images are compared by the machine learning module 135 with its training dataset to determine if the adverse event is about to happen. As soon as the machine learning module 135 determines that the adverse event is about to happen, the machine learning module 135 instructs the communication module 140a to transmit a control signal to the communication module 140b to switch off the power source to the heat source 115. Consequently, the food item being cooked is saved from being spoilt due to happening of an adverse action.

The detector module also includes an audio/visual alert module 150 to provide an audio and/or visual alert to a user of the system 100, thereby providing an alert in the event the cooking process needs to be stopped upon happening of the defined adverse event. The alert could either be in the form of an audio buzzer, alarm, LED or text display.

The present invention also envisages that a user may have control over the cooking process by configuring the detector module to monitor the cooking process as per his/her liking. For instance, the user may want the food to be optimally cooked to achieve a particular type of taste/texture. For example, while boiling potatoes in a pressure cooker, the user may want that the boiling process should only be carried out until three pressure release whistles of the pressure cooker happen. The user may thus instruct the detector module 120 accordingly using a user control module 155 configured on a user’s handheld device that is communicably coupled to the detector module 120. The user control module is capable of receiving inputs from the user regarding the desired parameters, such as number of pressure release whistles required or the type of texture, from the user to prevent happening of any adverse event in relation to the cooking process that the user wants the detector module to monitor and control.

It would thus be apparent that the present invention serves several advantages over the existing state of the art. In particular, any cooking process could be monitored and controlled without requiring any human intervention. More particularly, unlike the prior arts, the user is not required to continually monitor the food being cooked and the decision making regarding whether to continue or stop the cooking process is taken by the system of the present invention based on its analysis of the images of the food product during the course of the cooking process. This helps in saving time of the human user in monitoring the cooking process. The present invention could be largely beneficial in a place where large scale cooking process is being carried out, such as a restaurant, to cook food without much human involvement. As already stated, the present invention could be employed for any type of cooking process. The complete specification already described the system being employed for boiling of a liquid and cooking using pressure cooker. Other examples of cooking process where the system could be employed are, but not limited to, boiling eggs and popping popcorns.

It will be further appreciated that functions or structures of a plurality of components or steps may be combined into a single component or step, or the functions or structures of one-step or component may be split among plural steps or components. The present invention contemplates all of these combinations. Unless stated otherwise, dimensions and geometries of the various structures depicted herein are not intended to be restrictive of the invention, and other dimensions or geometries are possible. In addition, while a feature of the present invention may have been described in the context of only one of the illustrated embodiments, such feature may be combined with one or more other features of other embodiments, for any given application. It will also be appreciated from the above that the fabrication of the unique structures herein and the operation thereof also constitute methods in accordance with the present invention. The present invention also encompasses intermediate and end products resulting from the practice of the methods herein. The use of “comprising” or “including” also contemplates embodiments that “consist essentially of” or “consist of” the recited feature.

Although embodiments for the present invention have been described in language specific to structural features, it is to be understood that the present invention is not necessarily limited to the specific features described. Rather, the specific features and methods are disclosed as embodiments for the present invention. Numerous modifications and adaptations of the system/component of the present invention will be apparent to those skilled in the art, and thus it is intended by the appended claims to cover all such modifications and adaptations which fall within the scope of the present invention.
,CLAIMS:
1. A system to monitor and control adverse events in a cooking process of a food product, the system comprising:
a cooking vessel placed on a heat source for cooking the food product, the heat source being controlled by a controller unit; and
a detector module comprising
an imaging module placed proximal to the cooking vessel to capture images of the food product at pre-determined intervals within the cooking vessel during the course of the cooking process,
a processor communicably connected to the imaging module to receive the captured images of the food product, the processor comprising a machine learning module that evaluates the captured images of the food product based on a training dataset pertaining to the food product to determine the likelihood of happening of the adverse event,
a communication module communicably coupled to the processor and the controller unit fortransmitting a control signal to the controller unit upon the determination of likelihood of the adverse event by the processor, wherein the control signal switches off the heat source, to prevent the adverse event.

2. The system as claimed in claim 1, wherein the training dataset comprises at least hundred images of variations of the adverse event pertaining to the specific food item being cooked.

3. The system as claimed in claim 1, wherein the cooking process comprises boiling a liquid in the cooking vessel.

4. The system as claimed in claim 3, wherein the adverse event is overflowing of the liquid from the cooking vessel due to prolonged boiling.

5. The system as claimed in claim 1, wherein the cooking process comprises cooking a food product in a pressure cooker.

6. The system as claimed in claim 5, wherein the adverse event is the number of pressure release operation of the pressure cooker exceeding a specified number of pressure release operations.

7. The system as claimed in claim 1, wherein the imaging module comprises a digital camera.

8. The system as claimed in claim 7, wherein the digital camera comprises one of a standalone single wireless camera, a phone/tablet camera and a DSLR or a mirrorless camera.

9. The system as claimed in claim 1, wherein the controller unit is configured to regulate, switch on and switch off a power supply to the heat source to control operation thereof.

10. A method to monitor and control adverse events in a cooking process of a food product, the method comprising:
obtaining images of the food product at pre-determined intervalsduring the course of the cooking process;
evaluating the obtained images of the food product based on a training dataset pertaining to the food product to determine the likelihood of happening of the adverse event; and
transmitting a control signal upon the determination of the likelihood of the adverse eventto prevent the adverse event by stopping the cooking process,
wherein the training dataset comprises at least hundred images of variations of the adverse event pertaining to the specific food item being cooked.