Reinforcement Learning to Improve Color Adjustments in the Ceramic Industry

Proceedings Paper Computer Science, Artificial Intelligence

Improving Color Mixture Predictions in Ceramics using Data-centric Deep Learning

Tomas Souper et al.

Summary: This study explores the use of Deep Learning to generate color mixture predictions in ceramic glazes. By using spectral data and simulating the color mixing result digitally, a fully connected neural network model achieved the best performance. This approach shows potential for improving color mixing procedures in the ceramic industry.

PROCEEDINGS OF 2023 8TH INTERNATIONAL CONFERENCE ON MACHINE LEARNING TECHNOLOGIES, ICMLT 2023 (2023)

Add to Collection

Article Computer Science, Information Systems

Reinforcement Learning for Logic Recipe Generation: Bridging Gaps From Images to Plans

Mengyang Zhang et al.

Summary: A novel recipe generation system is proposed in this paper, which introduces ingredient generation to guide the production of effective recipes. A hierarchical attention mechanism is designed for feature extraction and a specific criterion is established to ensure the comprehensiveness and logic in the generated recipes.

IEEE TRANSACTIONS ON MULTIMEDIA (2022)

Add to Collection

Article Chemistry, Analytical

Color Design Decisions for Ceramic Products Based on Quantification of Perceptual Characteristics

Yi Wang et al.

Summary: This study proposes a method for quantifying ceramic color characteristics based on the BP neural network algorithm, which solves the mapping problem between the appearance characteristics of ceramic color and perceptual semantics, and provides a decision basis for ceramic product color design.

SENSORS (2022)

Add to Collection

Article Chemistry, Applied

Dyeing recipe prediction of cotton fabric based on hyperspectral colour measurement and an improved recurrent neural network

Jianxin Zhang et al.

Summary: A precise dyeing recipe prediction model for cotton fabric dyeing is proposed based on hyperspectral colour measurement and a deep learning algorithm. The model provides higher prediction accuracy for Reactive Red CI 238, Reactive Blue CI 204 and Reactive Yellow CI 206 compared to the Datacolor 650 recipe prediction model.

COLORATION TECHNOLOGY (2021)

Add to Collection

Proceedings Paper Computer Science, Theory & Methods

Model for Cooking Recipe Generation using Reinforcement Learning

Jumpei Fujita et al.

Summary: The study proposed a recipe-generation model in the encoder-decoder framework, which introduced reinforcement learning and coverage loss to improve the performance of recipe generation in matching input ingredients. Experimental results showed that the proposed model improved ingredient matching by approximately 21% compared to existing models.

2021 IEEE 37TH INTERNATIONAL CONFERENCE ON DATA ENGINEERING WORKSHOPS (ICDEW 2021) (2021)

Add to Collection

Article Engineering, Industrial