KR-20260063356-A - Identification system for plant-based meat substitutes with added root crops using visible and near-infrared hyperspectral imaging

KR20260063356AKR 20260063356 AKR20260063356 AKR 20260063356AKR-20260063356-A

Abstract

The present invention relates to a system for identifying plant-based alternative meat containing root crops, and more specifically, to a system for identifying the components of plant-based alternative meat containing root crops using a non-destructive method utilizing visible light and near-infrared hyperspectral imaging. According to one embodiment of the present invention, a system for identifying plant-based alternative meat containing root crops using visible light and near-infrared hyperspectral imaging is provided, wherein the identification system comprises: a hyperspectral image acquisition device for photographing a sample of plant-based alternative meat containing root crops; and a hyperspectral image analysis device comprising a component identification model that analyzes the hyperspectral image of the sample acquired by the hyperspectral image acquisition device to identify the components of the sample, and the component identification model is configured to identify the components of the alternative meat through a deep learning model or a machine learning model.

Inventors

노현권
이훈수

Assignees

충북대학교 산학협력단

Dates

Publication Date: 20260507
Application Date: 20241030

Claims (5)

In a system for identifying plant-based alternative meat containing root crops using visible light and near-infrared hyperspectral imaging, Hyperspectral imaging acquisition device for photographing a sample of plant-based alternative meat with added root crops; and A hyperspectral image analysis device comprising a component identification model that identifies the components of a sample by analyzing the hyperspectral image of the sample acquired from the hyperspectral image acquisition device; The above-mentioned component identification model is characterized by being configured to identify the components of alternative meat through a deep learning model or a machine learning model. A system for identifying plant-based alternative meat containing root crops using visible light and near-infrared hyperspectral imaging.
In paragraph 1, A deep learning training model or a machine learning training model is characterized by including any one of PLS-DA, Random Forest, SVM, KNN, Stacking, and CNN. A system for identifying plant-based alternative meat containing root crops using visible light and near-infrared hyperspectral imaging.
In paragraph 2, The above root crop is characterized by including any one of Gastrodia elata, Platycodon grandiflorus, Codonopsis pilosula, and ginseng. A system for identifying plant-based alternative meat containing root crops using visible light and near-infrared hyperspectral imaging.
In paragraph 2, When PLS-DA is used as a component discrimination model, the discrimination model operates to detect components of root crops contained in alternative meat based on wavelength bands including 469 nm, 554 nm, 609 nm, 633 nm, 726 nm, 752 nm, 797 nm, 921 nm, and 955 nm. A system for identifying plant-based alternative meat containing root crops using visible light and near-infrared hyperspectral imaging.
In paragraph 2, The spectrum used in visible and near-infrared hyperspectral imaging is characterized by being in the wavelength range of 400 nm to 1000 nm. A system for identifying plant-based alternative meat containing root crops using visible light and near-infrared hyperspectral imaging.

Description

Identification system for plant-based meat substitutes with added root crops using visible and near-infrared hyperspectral imaging The present invention relates to a system for identifying plant-based alternative meat containing root crops, and more specifically, to a system for identifying the components of plant-based alternative meat containing root crops using a non-destructive method utilizing visible light and near-infrared hyperspectral imaging. Since the 1960s, Korea's industrialization has progressed rapidly, leading to an improvement in national income. In fact, real per capita Gross National Income (GNI), which was 1.33 million won in the 1960s, increased to 37.77 million won in 2020. As national income improved, meat consumption also increased rapidly; it can be seen that per capita meat consumption, which was 27.45 kg in 1995, reached 57.40 kg in 2022. This increase in consumption led to an increase in demand. In 2006, there were 2,484 thousand cattle and 9,382 thousand pigs, but by 2023, these numbers had increased to 9,382 thousand and 11,089 thousand, respectively. As the number of livestock raised increases, the environmental impact of the livestock industry is growing. Traditional meat production methods consume large amounts of resources; to obtain 1 kg of meat, 7 to 10 kg of feed is used for cattle and 4 to 5 kg for pigs. (Opportunities and Challenges of Alternative Meat) In addition, it can be seen that greenhouse gas emissions from the livestock industry, which pollutes the atmosphere, increased by 77.2% from 5.8 million tons CO2eq. in the 1990s to 10.3 million tons CO2eq. in 2021. Various studies are being conducted both domestically and internationally to address these issues. Plant-based meat substitutes are garnering attention as one of the alternatives to solve these problems. The production method of plant-based meat substitutes is more efficient than conventional meat production methods. While beef shows a conversion rate of 10% to 14% and pork 20% to 25%, the resource conversion rate of plant-based meat substitutes is 75%. With its environmental advantages, meat substitutes are emerging as a solution to food shortages caused by the upcoming population growth. Furthermore, the alternative meat market is expanding as more people reduce their meat consumption for personal beliefs or ethical reasons. With the expansion of this market, various investments and research related to alternative meat are being conducted in Korea. In fact, research has been carried out to analyze the ingredients of commercially available alternative meats. Additionally, studies have been conducted to analyze the quality characteristics of alternative meat by adding mushrooms, which are functional foods. While various studies are being conducted by adding functional foods to alternative meat, research on alternative meat using root crops has not yet been carried out. Root crops have long been used as health foods in East Asia and are still consumed by many people today. Ginseng is widely consumed as a restorative food, while balloon flower root is consumed by those suffering from bronchial diseases such as the common cold. Research indicates that both of these crops have excellent antioxidant effects. Gastrodia elata possesses both antioxidant and anticancer properties. However, root crops can cause problems due to side effects. For example, there are cases of patients visiting the clinic after consuming raw ginseng, and surveys conducted on children revealed that some children had allergies to ginseng. Meat substitutes fortified with functional ingredients have positive effects in terms of function and nutrition. However, the potential for side effects should not be overlooked. To solve these problems, it is necessary to accurately identify the ingredients of the food being consumed. To determine the ingredients precisely, accurate analysis of the food must be performed. Various techniques have been used to analyze foods composed of a mixture of diverse ingredients. Representative methods include chemical methods and non-destructive methods using hyperspectral imaging. Conventional chemical analysis has the disadvantage of requiring sample destruction and taking a relatively long time. However, analysis using hyperspectral imaging has the advantage of being non-destructive as it preserves the original state of the sample, as well as the speed of analysis. Furthermore, chemical data can be obtained through spectral information along with physical data. For this reason, various studies utilizing hyperspectral imaging have been conducted recently. It has primarily been used to identify substances mixed into meats such as Halal, beef, pork, lamb, and chicken. Additionally, research has been conducted to identify mixtures of minced beef with meats such as pork, duck, and chicken. Moreover, while hyperspectral imaging has been used to analyze the components of alternative meats, research on non-destructively te