System and method for monitoring and quality evaluation of perishable food items

Patent metadata

Field	Value
Publication number	US-2020251229-A1
Application number	US-202016783897-A
Country	US
Kind code	A1
Filing date	Feb 6, 2020
Priority date	Feb 6, 2019
Publication date	Aug 6, 2020
Grant date	—

Abstract

This disclosure relates generally to a system and method for monitoring and quality evaluation of perishable food items in quantitative terms. Current technology provides limited capability for controlling environmental conditions surrounding the food items in real-time or any quantitative measurement for the degree of freshness of the perishable food items. The disclosed systems and methods facilitate in quantitative determination of freshness of food items by utilizing sensor data and visual data obtained by monitoring the food item. In an embodiment, the system utilizes a pre-trained CNN model and a RNN model, where the pertained CNN model is further fine-tined while training the RNN model to provide robust quality monitoring of the food items. In another embodiment, a rate kinetic based model is utilized for determining reaction rate order of the food item at a particular post-harvest stage of the food item so as to determine the remaining shelf life thereof.

First claim

What is claimed is:

1 . A processor implemented method, comprising:
obtaining a rate kinetic data associated with a food item enclosed in a networked framework, via one or more hardware processors, wherein the rate kinetic data comprises a time-series data having a plurality of attributes indicative of freshness of the food item; selectively partitioning the rate kinetic data into a plurality of post-harvest lifecycle stages of the food item based on a reaction rate order associated with one or more time intervals of each of the plurality of post-harvest lifecycle stages, via the one or more hardware processors, wherein the reaction rate order associated with the one or more time intervals of each of the plurality of post-harvest lifecycle stages is determined by a trained rate kinetic model; estimating, based at least on the reaction rate order associated with the one or more time intervals and the plurality of attributes, a plurality of values of shelf-life of the food item during each of the one or more time intervals, via the one or more hardware processors; aggregating a set of values of the shelf-life from amongst the plurality of values of the shelf-life corresponding to each attribute of the plurality of attributes, via the one or more hardware processors; and selecting, from amongst the set of values of the shelf-life, a minimum value of shelf-life as the food freshness value of the food item, via the one or more hardware processors.

2 . The method of claim 1 , wherein the rate kinetic data comprises at least a sensory data associated with the food item.

3 . The method of claim 1 , further comprising iteratively fitting training data on to reaction rate equations to obtain the trained rate kinetic model for the food item.

4 . The method of claim 1 , further comprising determining the plurality of post-harvest life cycle stages of the food item by selectively partitioning the input data into the plurality of post-harvest life cycle stages of the food item, wherein determining a post-harvest life cycle stage of the plurality of post-harvest life cycle stages comprises:
converting the time-series data into an integrated form of zero, first and second order rate reaction equations in a sequential order to obtain a plot of rate of change of concentration terms associated with the integral form of the trained rate kinetic model; obtaining a value of coefficient of determination from the plot of integral form of the zero, first and second order reactions in the sequential order; and determining the order of reaction based on a comparison of the value of the coefficient of determination with a threshold value of the coefficient of determination, wherein, the order of the reaction is indicative of a nature of reaction responsible for biochemical changes in the food item during the post-harvest lifecycle stage of the food item.

5 . The method of claim 4 , wherein the trained rate kinetic model for each of the plurality of post-harvest life cycle stages of the food item comprises one or more rate kinetic parameters, further wherein the one or more rate kinetic parameters comprises a Preexponential factor (KO) and an activation energy (E), further wherein the one or more rate kinetic parameters are calculated using the time-series data at a plurality of distinct temperatures.

6 . The method of claim 5 , wherein the trained rate kinetic model for each of the plurality of post-harvest life cycle stages of the food item comprises one or more parameters calculated using Arrhenius type equation, further wherein the Arrhenius type equation comprises parameters for capturing an effect of relative humidity on the attribute, further wherein the one or more parameters of the Arrhenius-type equation comprises a pre-exponential factor (K 0 ), a cultivar specific constant (m), and further wherein the one or more parameters are calculated using the time-series data at a plurality of distinct relative humidity values.

7 . The method of claim 1 , wherein the rate kinetic data comprises a sensory data associated with the food item, and wherein the rate kinetic data comprises weight of the food item, weight loss of the food item, moisture content in the food item, moisture loss during the storage, concentration of specific compound in the food item, and concentration of specific gas such as Carbon dioxide (CO2), ethylene (C2H4,), ammonia (NH3) released by the food item.

8 . The method of claim 6 , wherein the rate kinetic data further comprises visual data associated with the food item

9 . The method of claim 7 , wherein obtaining the rate kinetic data comprises collecting at least one of the visual data and the sensory data using at least one of invasive techniques and non-invasive techniques, the invasive techniques comprises use of one or more laboratory techniques to calculate food compositional parameters including at least one of sugar, starch, fat, protein, vitamins and antioxidants, and the non-invasive techniques comprises at least one of a plurality of non-invasive sensors, the plurality of non-invasive sensors comprises gas sensors, acoustics, optical sensors, and near infrared sensor.

10 . A system ( 300 ) comprising:
one or more memories ( 304 ); and one or more hardware processors ( 302 ), the one or more memories ( 304 ) coupled to the one or more hardware processors ( 302 ), wherein the one or more hardware processors ( 302 ) are configured to execute programmed instructions stored in the one or more memories ( 304 ), to: obtain a rate kinetic data associated with a food item enclosed in a networked framework, wherein the rate kinetic data comprises a time-series data having a plurality of attributes indicative of freshness of the food item; selectively partition the rate kinetic data into a plurality of post-harvest lifecycle stages of the food item based on a reaction rate order followed in one or more time intervals of each of the plurality of post-harvest lifecycle stages, wherein the reaction rate order associated with one or more time intervals of each of the plurality of post-harvest lifecycle stages is determined by a trained rate kinetic model; estimate, based at least on the reaction rate order associated with the one or more time intervals and the plurality of attributes, a plurality of values of shelf-life of the food item during each of the one or more time intervals; aggregate a set of values of the shelf-life from amongst the plurality of values of the shelf-life corresponding to each attribute of the plurality of attributes; and select, from amongst the set of values of the shelf-life, a minimum value of shelf-life as the food freshness value of the food item.

11 . The system of claim 10 , wherein the rate kinetic data comprises at least a sensory data associated with the food item.

12 . The system of claim 10 , wherein the one or more hardware processors are further configured to by the instructions to determine the plurality of post-harvest life cycle stages of the food item, and wherein to determine the plurality of post-harvest life cycle stages, the one or more hardware processors are further configured to by the instructions to selectively partition the input data into the plurality of post-harvest life cycle stages of the food item, further wherein to determine a post-harvest life cycle stage of the plurality of post-harvest life cycle stages, the one or more hardware processors are further configured to by the instructions to:
convert the time-series data into an integrated form of zero, first and second order rate reaction equations in a sequential order to obtain a plot of rate of change of concentration terms associated with t

Assignees

• Tata Consultancy Services Ltd

Inventors

• Rai Beena
• Dutta Jayita
• Deshpande Parijat
• Kausley Shankar Balajirao
• Karande Shirish Subhash
• Patwardhan Manasi Samarth
• Deshmukh Shashank Madhukar

Classifications

• G06V20/68

  Food, e.g. fruit or vegetables · CPC title

• G06V10/765

  using rules for classification or partitioning the feature space · CPC title

• G06V10/82

  using neural networks · CPC title

• G06V10/454

  Integrating the filters into a hierarchical structure, e.g. convolutional neural networks [CNN] · CPC title

• H04L67/01

  Protocols · CPC title