System and method for non-invasive real-time prediction of liquid food quality within enclosed package
State of the art food quality measurement techniques fail to determine quality of the food item once it is packed and sealed in a container. The disclosure herein generally relates to food quality prediction, and, more particularly, to a system and method for predicting food quality in a non-invasive manner. A Color Changing Indicator (CCI) in a biosensor strip forming a component of the enclosed package in which the liquid food item is packed, changes color when came in contact with the liquid food item. For different quality of the liquid food item the CCI has different color. Based on the color of the CCI, and ambient temperature and relative humidity at the time the color of the CCI is determined, a machine learning model determines rate of deterioration of the liquid food item, and then predicts remaining shelf life, which in turn provided as output to a user.
1. A method for non-invasive real-time prediction of quality of a liquid food item within enclosed package, comprising:
capturing, via one or more hardware processors, an image of a Color Changing Indicator (CCI) in a bio-sensor strip forming a component of the enclosed package, when the liquid food item comes in contact with the CCI, wherein the CCI comprises:
a transparent poly-di-methyl-siloxane (PDMS) substrate;
a thin film layer of bio-edible and bio-compatible color changing pigments, wherein the bio-edible and bio-compatible color changing pigments change color by interacting with one or more chemical components of the liquid food item, wherein a plurality of physio-thermal properties of each of the one or more chemical components vary with degradation of the liquid food item; and
an optical device, wherein the color change of the color changing pigments is visible through a transparent lens of the optical device;
determining, via the one or more hardware processors, a color of the CCI;
processing, via the one or more hardware processors using a machine learning data model, information on a) the determined color of the CCI, b) a measured ambient temperature inside the enclosed package while determining the color of the CCI and c) a measured relative humidity inside the enclosed package while determining the color of the CCI;
determining, via the one or more hardware processors, a value of a remaining shelf life of the liquid food item by:
determining a current quality of the liquid food item for the determined color of the CCI, the measured ambient temperature inside the enclosed package, and the measured relative humidity inside the enclosed package, in comparison with training data, wherein the training data used for training the machine learning data model comprises information on quality of the liquid food item corresponding to a plurality of combinations of a) a color of CCI, b) a value of the measured ambient temperature, and c) a value of the measured relative humidity;
determining a rate of deterioration of the liquid food item, based on the determined current quality of the liquid food item, and time expired, wherein the time expired is measured based on packaging date of the liquid food item; and
determining the value of the remaining shelf life of the liquid food item, based on the determined rate of deterioration; and
generating, via the one or more hardware processors, a result indicating the determined value of the remaining shelf life of the liquid food item.
2. A system for non-invasive real-time prediction of quality of a liquid food item within enclosed package, comprising:
one or more hardware processors;
a communication interface; and
a memory storing a plurality of instructions, wherein the plurality of instructions when executed, cause the one or more hardware processors to:
capture an image of a Color Changing Indicator (CCI) in a bio-sensor strip forming a component of the enclosed package, when the liquid food item comes in contact with the CCI, wherein the CCI comprises:
a transparent poly-di-methyl-siloxane (PDMS) substrate;
a thin film of bio-edible and bio-compatible color changing pigments, wherein the bio-edible and bio-compatible color changing pigments change color by interacting with one or more chemical components of the liquid food item, wherein a plurality of physio-thermal properties of each of the one or more chemical components vary with degradation of the liquid food item; and
an optical device, wherein the color change of the color changing pigments is visible through a transparent lens of the optical device;
process, using a machine learning data model, information on a) the determined color of the CCI, b) a measured ambient temperature inside the enclosed package while determining the color of the CCI, and c) a measured relative humidity inside the enclosed package while determining the color of the CCI;
determine a value of a remaining shelf life of the liquid food item by:
determining a current quality of the liquid food item for the determined color of the CCI, the measured ambient temperature inside the enclosed package, and the measured relative humidity inside the enclosed package, in comparison with training data, wherein the training data used for training the machine learning data model comprises information on quality of the liquid food item corresponding to a plurality of combinations of a) a color of CCI, b) a value of the measured ambient temperature, and c) a value of the measured relative humidity;
determining a rate of deterioration of the liquid food item, based on the determined current quality of the liquid food item, and time expired, wherein the time expired is measured based on packaging date of the liquid food item; and
determining the value of the remaining shelf life of the liquid food item, based on the determined rate of deterioration; and
generate a result indicating the determined value of the remaining shelf life of the liquid food item.
3. A non-transitory computer readable medium for non-invasive real-time prediction of quality of a liquid food item within enclosed package, wherein the non-transitory computer readable medium comprising a plurality of instructions, which when executed, cause:
capturing an image determining color of a Color Changing Indicator (CCI) in a bio-sensor strip forming a component of the enclosed package, when the liquid food item comes in contact with the CCI, via one or more hardware processors, wherein the CCI comprises:
a transparent poly-di-methyl-siloxane (PDMS) substrate;
a thin film layer of bio-edible and bio-compatible color changing pigments, wherein the bio-edible and bio-compatible color changing pigments change color by interacting with one or more chemical components of the liquid food item, wherein a plurality of physio-thermal properties of each of the one or more chemical components vary with degradation of the liquid food item; and
an optical device, wherein the color change of the color changing pigments is visible through a transparent lens of the optical device;
processing, using a machine learning data model, information on a) the determined color of the CCI, b) a measured ambient temperature inside the enclosed package while determining the color of the CCI, and c) a measured relative humidity inside the enclosed package while determining the color of the CCI;
determining a value of a remaining shelf life of the liquid food item by:
determining a current quality of the liquid food item for the determined color of the CCI, the measured ambient temperature inside the enclosed package, and the measured relative humidity inside the enclosed package, in comparison with training data, wherein the training data used for training the machine learning data model comprises information on quality of the liquid food item corresponding to a plurality of combinations of a) a color of CCI, b) a value of the measured ambient temperature, and c) a value of the measured relative humidity;
determining a rate of deterioration of the liquid food item, based on the determined current quality of the liquid food item, and time expired, wherein the time expired is measured based on packaging date of the liquid food item; and
determining the value of the remaining shelf life of the liquid food item, based on the determined rate of deterioration; and
generating a result indicating the determined value of the remaining shelf life of the liquid food item.