Hexa Food’s IoT team has deployed Huawei’s ModelArts coupled with the intelligent device Atlas 500 to accurately identify the quality of the chilies it uses in spice blends. The AI can distinguish good chilies from bad, improving production efficiency and the quality of spices available to chefs and homes across Malaysia.
Known as the “Kingdom of Spices”, Malaysia is a multi-ethnic nation comprising Malays, Chinese, Indians, and the indigenous Orang Asli people. Its diverse culture is reflected in its cuisine, which draws from a multicultural heritage that sees hundreds of spices add flavor to the Malaysian diet. And of these, the colorful, aromatic, and spicy chili powder is a mainstay of many of the nation’s signature dishes.
Sorting the good from the bad
The quality of chilies determines the flavor of the spice blends that produce chili powder. Hexa Food believes that good spice blends can only be made from outstanding ingredients, so sorting raw materials is a priority for the company.
Hexa Food has been producing herbs, spices, and seasonings since 2007. Staff spend much of their time attempting to differentiate between good and bad chilies, selecting only the best ones to add to spice blends. However, sorting chilies like this is both labor- and time-intensive and, with subjectivity in the mix, leads to inconsistent quality.
AI: The new expert
To remain competitive, Hexa Food has already embarked on the digital transformation journey, setting up an online concept store and establishing an IoT team to study how to apply AI, big data, and IoT to intelligent production. The IoT team has built and trained a chili identification model on the ModelArts platform, with the Atlas 500 deployed at the edge. The Atlas 500’s image recognition technologies can quickly and accurately identify chili quality, distinguishing good from bad, and improving production efficiency and the quality of spice blends. The solution provides the following functions:
- Provides intelligent AI-powered sorting, which eliminates errors in manual sorting and improves efficiency by 50 percent. With the auto-learning capability of ModelArts, developers with little coding or model development experience only have to perform one step in the AI development process.
- Simultaneously processes 16 video streams at the edge to analyze chili appearance and quality. The Atlas 500 integrates multiple data interfaces and can achieve edge-cloud synergy with Huawei Cloud. The training models are sent from the cloud and updated at the edge in real time.
For Hexa Food, the AI distills human experience and the traditional recipes of Malaysia's unique food culture to ensure maximum quality. Currently, Hexa Food provides high-quality spices for customers such as Giant, Tesco, AEON BiG, LIFE, and MISSION, and as a result of its noted quality, recently won second place in Malaysia’s prestigious Enterprise 50 (E50) awards.
AI will continue to ensure that the company remains a major player in Malaysian cuisine.