Abstract
In this paper, a new image analysis method based on an in-situ microscopic imaging system is proposed for detecting micro crystals in cooling crystallization. Due to the limitation of measurement technology, it is a challenge to extract the evolutionary information of micro crystals, which are too small to be precisely analyzed by in-situ images, e.g. crystals at the initial crystallization stage. An improved deep-learning model is used to enhance the image resolution of micro crystals, thus more effectively obtaining the crystal shape and size information. In addition, a valid size calibration method by simulating particle motion is proposed. Consequently, image size measurement can be easily performed for crystals by using an axis-based algorithm. Experimental verifications on $\beta $ -form L-glutamic acid crystallization were performed to demonstrate the effectiveness of the proposed method for detecting micro crystal information.
Highlights
With the fast progress of process analytical technology (PAT) for materials design, chemical engineering etc., on-line imaging systems with microscopic camera sensors have been performed to detect crystal size, shape, degree of agglomeration and growth rate in industrial crystallization processes in the recent years [1], [2]
To make a quantitative analysis in this work, five off-line microscope images were used in the same experimental environment to test quantitatively the effectiveness of improved algorithm, compared with Bicubic interpolation, SRCNN [21], VDSR [22] and residual dense network (RDN) [20] in crystal image super-resolution
An in-situ image analysis method has been proposed for estimation of micro crystal shape and size at initial L-glutamic acid (LGA) crystallization stage
Summary
With the fast progress of process analytical technology (PAT) for materials design, chemical engineering etc., on-line imaging systems with microscopic camera sensors have been performed to detect crystal size, shape, degree of agglomeration and growth rate in industrial crystallization processes in the recent years [1], [2]. Recent references studied the measurement and classification methods of crystal images, which realized the analysis of crystal evolution based on on-line imaging [8]-[11]. Manee et al [13] proposed a new detection method based on deeplearning model for crystal images in high-density slurry. Wu et al [14] developed a state-of-the-art deeplearning model to perform size analysis by using the costeffective imaging system consisting of a flow-through cell for real-time imaging
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call