基于原位成像与图像分析技术的结晶过程智能监测

doi:10.11949/0438-1157.20250104

摘要/Abstract

摘要：

结晶是物质分离与纯化的重要技术，目前主要采用离线分析的方法监测结晶过程参数，存在滞后性、消耗人力等弊端。基于过程分析技术（process analytical technology，PAT）的在线监测对于结晶生产工艺的优化与控制具有重要意义，可以充分提高生产效率与产品质量。原位成像技术能直观地反映结晶状态，图像分析算法提供的定量信息可用于研究结晶动力学等机理。虽然原位图像分析极具挑战性，但近些年以深度学习为代表的人工智能技术显著提升了预测精度。在简要介绍结晶领域各种常用PAT技术的基础上，重点综述了基于原位成像技术的相关研究进展，侧重于图像分析方法及其揭示的结晶行为与机理。最后，展望了原位成像技术对实现结晶过程智能监测的一些未来发展方向。

关键词: 结晶, 成像, 粒度分布, 算法, 深度学习, 过程分析技术

Abstract:

Crystallization is an important technology for the separation and purification of materials. Currently, the parameters of crystallization processes are mainly monitored by off-line methods, which have some disadvantages, such as time lag and labor consumption. On-line monitoring based on process analytical technology (PAT) is of great significance for the optimization and control of crystallization, which can improve its production efficiency and product quality. In situ imaging technology can intuitively reflect the crystallization state, and the quantitative information provided by the image analysis algorithm can be used to study the mechanism such as crystallization kinetics. Although the analysis of in situ images is quite challenging, the recent AI technologies represented by deep learning have significantly enhanced the analysis accuracy. Following a brief introduction of some common PAT techniques for crystallization monitoring, this review summarizes the progress on in situ imaging technique with focus on the image analysis methods together with the revealed behaviors and mechanisms of crystallization. Finally, some future directions of in situ imaging technology are prospected for the intelligent monitoring of crystallization processes.

Key words: crystallization, imaging, size distribution, algorithm, deep learning, process analytical technology

中图分类号:

TQ 014

周光正, 钟子翰, 黄彦群, 王学重. 基于原位成像与图像分析技术的结晶过程智能监测[J]. 化工学报, 2025, 76(9): 4351-4368.

Guangzheng ZHOU, Zihan ZHONG, Yanqun HUANG, Xuezhong WANG. Intelligent monitoring of crystallization processes based on in situ imaging and image analysis[J]. CIESC Journal, 2025, 76(9): 4351-4368.

图/表 10

参考文献 111

[1]	Cote A, Erdemir D, Girard K P, et al. Perspectives on the current state, challenges, and opportunities in pharmaceutical crystallization process development[J]. Crystal Growth & Design, 2020, 20(12): 7568-7581.
[2]	Jia S Z, Wan X X, Yao T, et al. Separation performance and agglomeration behavior analysis of solution crystallization in food engineering[J]. Food Chemistry, 2023, 419: 136051.
[3]	黄炎, 孙海龙, 孟子超, 等. 溶析结晶在医药领域的研究进展[J]. 化工进展, 2019, 38(5): 2380-2388.
	Huang Y, Sun H L, Meng Z C, et al. Progress in antisolvent crystallization in pharmaceutical field[J]. Chemical Industry and Engineering Progress, 2019, 38(5): 2380-2388.
[4]	Pu S Y, Hadinoto K. Continuous crystallization as a downstream processing step of pharmaceutical proteins: a review[J]. Chemical Engineering Research and Design, 2020, 160: 89-104.
[5]	Zhou H Y, Zhou G Z, Wang X Z. Multi-objective optimization of protein cooling crystallization with morphological population balance models[J]. Journal of Crystal Growth, 2022, 588: 126664.
[6]	Zhang D J, Xu S J, Du S C, et al. Progress of pharmaceutical continuous crystallization[J]. Engineering, 2017, 3(3): 354-364.
[7]	Orehek J, Teslić D, Likozar B. Continuous crystallization processes in pharmaceutical manufacturing: a review[J]. Organic Process Research & Development, 2021, 25(1): 16-42.
[8]	赵绍磊, 王耀国, 张腾, 等. 制药结晶中的先进过程控制[J]. 化工学报, 2020, 71(2): 459-474.
	Zhao S L, Wang Y G, Zhang T, et al. Advanced process control of pharmaceutical crystallization[J]. CIESC Journal, 2020, 71(2): 459-474.
[9]	Food and Drug Administration. Guidance for industry: PAT—a framework for innovative pharmaceutical development, manufacturing and quality assurance[R]. Maryland: FDA, 2004.
[10]	Grangeia H B, Silva C, Simões S P, et al. Quality by design in pharmaceutical manufacturing: a systematic review of current status, challenges and future perspectives[J]. European Journal of Pharmaceutics and Biopharmaceutics, 2020, 147: 19-37.
[11]	Simon L L, Pataki H, Marosi G, et al. Assessment of recent process analytical technology (PAT) trends: a multiauthor review[J]. Organic Process Research & Development, 2015, 19(1): 3-62.
[12]	张妍, 程景才, 杨超, 等. 药物多晶型的过程控制和工程技术进展[J]. 中国医药工业杂志, 2018, 49(5): 537-546.
	Zhang Y, Cheng J C, Yang C, et al. Progress in the process control of pharmaceutical polymorphism and engineering[J]. Chinese Journal of Pharmaceuticals, 2018, 49(5): 537-546.
[13]	Ma Y M, Wu S G, Macaringue E G J, et al. Recent progress in continuous crystallization of pharmaceutical products: precise preparation and control[J]. Organic Process Research & Development, 2020, 24(10): 1785-1801.
[14]	Nagy Z K, Braatz R D. Advances and new directions in crystallization control[J]. Annual Review of Chemical and Biomolecular Engineering, 2012, 3: 55-75.
[15]	龚俊波, 孙杰, 王静康. 面向智能制造的工业结晶研究进展[J]. 化工学报, 2018, 69(11): 4505-4517.
	Gong J B, Sun J, Wang J K. Research progress of industrial crystallization towards intelligent manufacturing[J]. CIESC Journal, 2018, 69(11): 4505-4517.
[16]	Zhang F K, Meng A Z, Long Y, et al. Advances and opportunities concerning nucleation measurement and control technology in crystallization[J]. Organic Process Research & Development, 2024, 28(8): 3055-3077.
[17]	Lu M J, Rao S L, Yue H, et al. Recent advances in the application of machine learning to crystal behavior and crystallization process control[J]. Crystal Growth & Design, 2024, 24(12): 5374-5396.
[18]	Zhu Z X, Zhang Y, Wang Z X, et al. Artificial intelligence assisted pharmaceutical crystallization[J]. Crystal Growth & Design, 2024, 24(10): 4245-4270.
[19]	Cornel J, Lindenberg C, Mazzotti M. Quantitative application of in situ ATR-FTIR and Raman spectroscopy in crystallization processes[J]. Industrial & Engineering Chemistry Research, 2008, 47(14): 4870-4882.
[20]	Liu X J, Zhang Y, Wang X Z. Study on co-crystallization of LCZ696 using in situ ATR-FTIR and imaging[J]. Crystals, 2020, 10(10): 922.
[21]	Hansen T B, Simone E, Nagy Z, et al. Process analytical tools to control polymorphism and particle size in batch crystallization processes[J]. Organic Process Research & Development, 2017, 21(6): 855-865.
[22]	Tacsi K, Gyürkés M, Csontos I, et al. Polymorphic concentration control for crystallization using Raman and attenuated total reflectance ultraviolet visible spectroscopy[J]. Crystal Growth & Design, 2020, 20(1): 73-86.
[23]	Mehmood T, Ahmed B. The diversity in the applications of partial least squares: an overview[J]. Journal of Chemometrics, 2016, 30(1): 4-17.
[24]	Zhang F K, Du K, Guo L Y, et al. Progress, problems, and potential of technology for measuring solution concentration in crystallization processes[J]. Measurement, 2022, 187: 110328.
[25]	Zhang Y, Zhang L, Zhou G Z, et al. PAT aided feasibility study on continuous crystallization of benzotriazole[J]. Organic Process Research & Development, 2024, 28(9): 3625-3636.
[26]	徐啟蕾, 郭鲁钰, 杜康, 等. 基于ATR-FTIR光谱测量结晶过程溶液浓度的变量稳定加权混合收缩方法[J]. 光谱学与光谱分析, 2023, 43(5): 1413-1418.
	Xu Q L, Guo L Y, Du K, et al. A hybrid shrinkage strategy based on variable stable weighted for solution concentration measurement in crystallization via ATR-FTIR spectroscopy[J]. Spectroscopy and Spectral Analysis, 2023, 43(5): 1413-1418.
[27]	Abu Bakar M R, Nagy Z K, Saleemi A N, et al. The impact of direct nucleation control on crystal size distribution in pharmaceutical crystallization processes[J]. Crystal Growth & Design, 2009, 9(3): 1378-1384.
[28]	Acevedo D, Wu W, Yang X C, et al. Evaluation of focused beam reflectance measurement (FBRM) for monitoring and predicting the crystal size of carbamazepine in crystallization processes[J]. CrystEngComm, 2021, 23(4): 972-985.
[29]	Agimelen O S, Svoboda V, Ahmed B, et al. Multi-sensor inline measurements of crystal size and shape distributions during high shear wet milling of crystal slurries[J]. Advanced Powder Technology, 2018, 29(12): 2987-2995.
[30]	De Anda J C, Wang X Z, Lai X, et al. Real-time product morphology monitoring in crystallization using imaging technique[J]. AIChE Journal, 2005, 51(5): 1406-1414.
[31]	Ma C Y, Liu J J, Wang X Z. Measurement, modelling, and closed-loop control of crystal shape distribution: literature review and future perspectives[J]. Particuology, 2016, 26: 1-18.
[32]	Simone E, Saleemi A N, Nagy Z K. Application of quantitative Raman spectroscopy for the monitoring of polymorphic transformation in crystallization processes using a good calibration practice procedure[J]. Chemical Engineering Research and Design, 2014, 92(4): 594-611.
[33]	Hu Y R, Liang J K, Myerson A S, et al. Crystallization monitoring by Raman spectroscopy: simultaneous measurement of desupersaturation profile and polymorphic form in flufenamic acid systems[J]. Industrial & Engineering Chemistry Research, 2005, 44(5): 1233-1240.
[34]	Simone E, Saleemi A N, Tonnon N, et al. Active polymorphic feedback control of crystallization processes using a combined Raman and ATR-UV/Vis spectroscopy approach[J]. Crystal Growth & Design, 2014, 14(4): 1839-1850.
[35]	Powell K A, Saleemi A N, Rielly C D, et al. Monitoring continuous crystallization of paracetamol in the presence of an additive using an integrated PAT array and multivariate methods[J]. Organic Process Research & Development, 2016, 20(3): 626-636.
[36]	Wilkinson M J, Jennings K H, Hardy M. Non-invasive video imaging for interrogating pharmaceutical crystallization processes[J]. Microscopy and Microanalysis, 2000, 6(S2): 996-997.
[37]	Blandin A F, Rivoire A, Mangin D, et al. Using in situ image analysis to study the kinetics of agglomeration in suspension[J]. Particle & Particle Systems Characterization, 2000, 17(1): 16-20.
[38]	Huo Y, Liu T, Liu H, et al. In-situ crystal morphology identification using imaging analysis with application to the L-glutamic acid crystallization[J]. Chemical Engineering Science, 2016, 148: 126-139.
[39]	Zhu Q H, Zhou G Z, Hou G H, et al. On-line image analysis for evaporative crystallization of xylose[J]. Powder Technology, 2025, 452: 120446.
[40]	Su W Y, Hao H X, Barrett M, et al. The impact of operating parameters on the polymorphic transformation of D-mannitol characterized in situ with Raman spectroscopy, FBRM, and PVM[J]. Organic Process Research & Development, 2010, 14(6): 1432-1437.
[41]	Liu W J, Wei H Y, Zhao J T, et al. Investigation into the cooling crystallization and transformations of carbamazepine using in situ FBRM and PVM[J]. Organic Process Research & Development, 2013, 17(11): 1406-1412.
[42]	Barrett P, Glennon B. Characterizing the metastable zone width and solubility curve using Lasentec FBRM and PVM[J]. Chemical Engineering Research and Design, 2002, 80(7): 799-805.
[43]	Liu J J, Ma C Y, Wang X Z. Imaging protein crystal growth behaviour in batch cooling crystallisation[J]. Chinese Journal of Chemical Engineering, 2016, 24(1): 101-108.
[44]	Tang X H, Liu J J, Zhang Y, et al. Study on the influence of lysozyme crystallization conditions on crystal properties in crystallizers of varied sizes when temperature is the manipulated variable[J]. Journal of Crystal Growth, 2018, 498: 186-196.
[45]	Jiang M, Braatz R D. Designs of continuous-flow pharmaceutical crystallizers: developments and practice[J]. CrystEngComm, 2019, 21(23): 3534-3551.
[46]	Domokos A, Nagy B, Szilágyi B, et al. Integrated continuous pharmaceutical technologies—a review[J]. Organic Process Research & Development, 2021, 25(4): 721-739.
[47]	Ferguson S, Morris G, Hao H X, et al. In-situ monitoring and characterization of plug flow crystallizers[J]. Chemical Engineering Science, 2012, 77: 105-111.
[48]	Su Q L, Nagy Z K, Rielly C D. Pharmaceutical crystallisation processes from batch to continuous operation using MSMPR stages: modelling, design, and control[J]. Chemical Engineering and Processing: Process Intensification, 2015, 89: 41-53.
[49]	Eren A, Civati F, Ma W C, et al. Continuous crystallization and its potential use in drug substance manufacture: a review[J]. Journal of Crystal Growth, 2023, 601: 126958.
[50]	Wang X Z, Roberts K J, Ma C Y. Crystal growth measurement using 2D and 3D imaging and the perspectives for shape control[J]. Chemical Engineering Science, 2008, 63(5): 1173-1184.
[51]	Schorsch S, Vetter T, Mazzotti M. Measuring multidimensional particle size distributions during crystallization[J]. Chemical Engineering Science, 2012, 77: 130-142.
[52]	Wu Y Y, Gao Z G, Rohani S. Deep learning-based oriented object detection for in situ image monitoring and analysis: a process analytical technology (PAT) application for taurine crystallization[J]. Chemical Engineering Research and Design, 2021, 170: 444-455.
[53]	Huo Y, Li X, Tu B B. Image measurement of crystal size growth during cooling crystallization using high-speed imaging and a U-Net network[J]. Crystals, 2022, 12(12): 1690.
[54]	Li Y, Zhang Y, Wang X Z. Secondary nucleation kinetics of AIBN crystallisation in methanol: online imaging-based measurement and modelling[J]. Crystals, 2020, 10(6): 506.
[55]	Patience D B, Rawlings J B. Particle-shape monitoring and control in crystallization processes[J]. AIChE Journal, 2001, 47(9): 2125-2130.
[56]	De Anda J C, Wang X Z, Roberts K J. Multi-scale segmentation image analysis for the in-process monitoring of particle shape with batch crystallisers[J]. Chemical Engineering Science, 2005, 60(4): 1053-1065.
[57]	Li C H, Lee C K. Minimum cross entropy thresholding[J]. Pattern Recognition, 1993, 26(4): 617-625.
[58]	Vancleef A, Maes D, Van Gerven T, et al. Flow-through microscopy and image analysis for crystallization processes[J]. Chemical Engineering Science, 2022, 248: 117067.
[59]	Nazeran H, Rice F, Moran W, et al. Biomedical image processing in pathology: a review[J]. Australasian Physical and Engineering Sciences in Medicine, 1995, 18(1): 26-38.
[60]	Kacker R, Maaß S, Emmerich J, et al. Application of inline imaging for monitoring crystallization process in a continuous oscillatory baffled crystallizer[J]. AIChE Journal, 2018, 64(7): 2450-2461.
[61]	LeCun Y, Bengio Y, Hinton G. Deep learning[J]. Nature, 2015, 521: 436-444.
[62]	Abramson J, Adler J, Dunger J, et al. Accurate structure prediction of biomolecular interactions with AlphaFold 3[J]. Nature, 2024, 630: 493-500.
[63]	He K M, Zhang X Y, Ren S Q, et al. Deep residual learning for image recognition[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas, NV, USA: IEEE, 2016: 770-778.
[64]	Dosovitskiy A. An image is worth 16×16 words: transformers for image recognition at scale[EB/OL]. [2025-02-03]. .
[65]	Ren S Q, He K M, Girshick R, et al. Faster R-CNN: towards real-time object detection with region proposal networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(6): 1137-1149.
[66]	Redmon J, Farhadi A. YOLOv3: an incremental improvement[EB/OL]. [2025-02-03]. .
[67]	He K M, Gkioxari G, Dollár P, et al. Mask R-CNN[C]//2017 IEEE International Conference on Computer Vision (ICCV). Venice, Italy: IEEE, 2017: 2980-2988.
[68]	Bolya D, Zhou C, Xiao F Y, et al. YOLACT: real-time instance segmentation[C]//2019 IEEE/CVF International Conference on Computer Vision (ICCV). Seoul, Korea: IEEE, 2019: 9156-9165.
[69]	Ronneberger O, Fischer P, Brox T. U-Net: convolutional networks for biomedical image segmentation[C]//Medical Image Computing and Computer-Assisted Intervention-MICCAI 2015. Cham: Springer International Publishing, 2015: 234-241.
[70]	Badrinarayanan V, Kendall A, Cipolla R. SegNet: a deep convolutional encoder-decoder architecture for image segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(12): 2481-2495.
[71]	Bruno A E, Charbonneau P, Newman J, et al. Classification of crystallization outcomes using deep convolutional neural networks[J]. PLoS One, 2018, 13(6): e0198883.
[72]	Milne J, Qian C, Hargreaves D, et al. Not getting in too deep: a practical deep learning approach to routine crystallisation image classification[J]. PLoS One, 2023, 18(3): e0282562.
[73]	Huang L, Yang D Y, Yu Z M, et al. Deep learning-aided high-throughput screening of time-resolved protein crystallization on programmable microliter-droplet systems[J]. Chemical Engineering Journal, 2022, 450: 138267.
[74]	Zhang J L, Meng Y M, Wu J F, et al. Monitoring sugar crystallization with deep neural networks[J]. Journal of Food Engineering, 2020, 280: 109965.
[75]	Salami H, McDonald M A, Bommarius A S, et al. In situ imaging combined with deep learning for crystallization process monitoring: application to cephalexin production[J]. Organic Process Research & Development, 2021, 25(7): 1670-1679.
[76]	Ding N, Bao X, Sun S T, et al. High-precision real-time urine crystallization recognition based on dilated bilinear space pyramid ConvNext[J]. International Journal of Imaging Systems and Technology, 2024, 34(2): e22999.
[77]	Minaee S, Boykov Y, Porikli F, et al. Image segmentation using deep learning: a survey[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 44(7): 3523-3542.
[78]	Gu W C, Bai S, Kong L X. A review on 2D instance segmentation based on deep neural networks[J]. Image and Vision Computing, 2022, 120: 104401.
[79]	Lin T Y, Dollár P, Girshick R, et al. Feature pyramid networks for object detection[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Honolulu, HI, USA: IEEE, 2017: 936-944.
[80]	Gao Z G, Wu Y Y, Bao Y, et al. Image analysis for in-line measurement of multidimensional size, shape, and polymorphic transformation of L-glutamic acid using deep learning-based image segmentation and classification[J]. Crystal Growth & Design, 2018, 18(8): 4275-4281.
[81]	Fang L, Liu J, Han D D, et al. Revealing the role of polymer in the robust preparation of the 2,4-dichlorophenoxyacetic acid metastable crystal form by AI-based image analysis[J]. Powder Technology, 2023, 413: 118077.
[82]	Zong S L, Zhou G Z, Li M, et al. Deep learning-based on-line image analysis for continuous industrial crystallization processes[J]. Particuology, 2023, 74: 173-183.
[83]	Li M Y, Liu J, Yao T, et al. Deep-learning based in situ micrograph analysis of high-density crystallization slurry using image and data enhancement strategy[J]. Powder Technology, 2024, 437: 119582.
[84]	Liu S, Qi L, Qin H F, et al. Path aggregation network for instance segmentation[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City, UT, USA: IEEE, 2018: 8759-8768.
[85]	He J X, Zhou J H, Dong J P, et al. Revealing the effects of microwell sizes on the crystal growth kinetics of active pharmaceutical ingredients by deep learning[J]. Chemical Engineering Journal, 2022, 428: 131986.
[86]	Manee V, Zhu W, Romagnoli J A. A deep learning image-based sensor for real-time crystal size distribution characterization[J]. Industrial & Engineering Chemistry Research, 2019, 58(51): 23175-23186.
[87]	Li X T, Ding H H, Yuan H B, et al. Transformer-based visual segmentation: a survey[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2024, 46(12): 10138-10163.
[88]	Vaswani A, Shazeer N, Parmar N, et al. Attention is all you need[C]//Advances in Neural Information Processing Systems. Curran Associates Inc., 2017: 6000-6010.
[89]	Zuo S Q, Xiao Y, Chang X J, et al. Vision transformers for dense prediction: a survey[J]. Knowledge-Based Systems, 2022, 253: 109552.
[90]	Liu Z, Lin Y T, Cao Y, et al. Swin transformer: hierarchical vision transformer using shifted windows[C]//2021 IEEE/CVF International Conference on Computer Vision (ICCV). Montreal, QC, Canada: IEEE, 2021: 9992-10002.
[91]	Xiao T T, Liu Y C, Zhou B L, et al. Unified perceptual parsing for scene understanding[C]// Computer Vision-ECCV 2018. Cham: Springer International Publishing, 2018: 432-448.
[92]	Wang H, Fan J, Liu T, et al. Deep learning-based image analysis with RTFormer network for measuring 2D crystal size distribution during cooling crystallization of β form L-glutamic acid[J]. Measurement, 2025, 242: 116227.
[93]	Zhang H W, Zhu Y, Wang D, et al. A survey on visual mamba[J]. Applied Sciences, 2024, 14(13): 5683.
[94]	Chlap P, Min H, Vandenberg N, et al. A review of medical image data augmentation techniques for deep learning applications[J]. Journal of Medical Imaging and Radiation Oncology, 2021, 65(5): 545-563.
[95]	Zhuang F Z, Qi Z Y, Duan K Y, et al. A comprehensive survey on transfer learning[J]. Proceedings of the IEEE, 2021, 109(1): 43-76.
[96]	Li R F, Thomson G B, White G, et al. Integration of crystal morphology modeling and on-line shape measurement[J]. AIChE Journal, 2006, 52(6): 2297-2305.
[97]	Borchert C, Temmel E, Eisenschmidt H, et al. Image-based in situ identification of face specific crystal growth rates from crystal populations[J]. Crystal Growth & Design, 2014, 14(3): 952-971.
[98]	Kempkes M, Vetter T, Mazzotti M. Measurement of 3D particle size distributions by stereoscopic imaging[J]. Chemical Engineering Science, 2010, 65(4): 1362-1373.
[99]	Schorsch S, Ochsenbein D R, Vetter T, et al. High accuracy online measurement of multidimensional particle size distributions during crystallization[J]. Chemical Engineering Science, 2014, 105: 155-168.
[100]	Ma C Y, Liu J J, Wang X Z. Stereo imaging of crystal growth[J]. AIChE Journal, 2016, 62(1): 18-25.
[101]	Zhang R, Ma C Y, Liu J J, et al. On-line measurement of the real size and shape of crystals in stirred tank crystalliser using non-invasive stereo vision imaging[J]. Chemical Engineering Science, 2015, 137: 9-21.
[102]	Wu K, Ma C Y, Liu J J, et al. Measurement of crystal face specific growth kinetics[J]. Crystal Growth & Design, 2016, 16(9): 4855-4868.
[103]	Huo Y, Liu T, Wang X Z, et al. Online detection of particle agglomeration during solution crystallization by microscopic double-view image analysis[J]. Industrial & Engineering Chemistry Research, 2017, 56(39): 11257-11269.
[104]	Huo Y, Liu T, Yang Y X, et al. In situ measurement of 3D crystal size distribution by double-view image analysis with case study onl-glutamic acid crystallization[J]. Industrial & Engineering Chemistry Research, 2020, 59(10): 4646-4658.
[105]	Chen Y B, Mancini M, Zhu X T, et al. Semi-supervised and unsupervised deep visual learning: a survey[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2024, 46(3): 1327-1347.
[106]	Shu Y D, Liu J J, Zhang Y, et al. Considering nucleation, breakage and aggregation in morphological population balance models for crystallization processes[J]. Computers & Chemical Engineering, 2020, 136: 106781.
[107]	周光正, 王学重, 周浩宇. 溶酶菌蛋白质结晶的多目标优化与模拟[J]. 化工学报, 2023, 74(10): 4191-4200.
	Zhou G Z, Wang X Z, Zhou H Y. Multi-objective optimization and simulation of lysozyme protein crystallization[J]. CIESC Journal, 2023, 74(10): 4191-4200.
[108]	李川, 洪振取, 单宝明, 等. 求解多维粒数衡算方程的高阶紧致差分方法[J]. 化工学报, 2024, 75(12): 4513-4522.
	Li C, Hong Z Q, Shan B M, et al. High-order compact difference method for solving the multidimensional population balance equation[J]. CIESC Journal, 2024, 75(12): 4513-4522.
[109]	Wang H L, Yang L, Li X Y, et al. Polydisperse particle inline image method and its application onto gas-liquid flow in a stirred tank[J]. AIChE Journal, 2024, 70(6): e18398.
[110]	Wang X L, Zhou G Z, Liang L P, et al. Deep learning-based image analysis for in situ microscopic imaging of cell culture process[J]. Engineering Applications of Artificial Intelligence, 2024, 129: 107621.
[111]	Zhang S Y, Liang X, Huang X Y, et al. Precise and fast microdroplet size distribution measurement using deep learning[J]. Chemical Engineering Science, 2022, 247: 116926.

视觉任务	功能	数据标签	代表算法
图像分类	将图像分到预定义类别中的某一类	说明图像所属的类别；工作量少	ResNet^[63] Vision Transformer^[64]
目标检测	识别所有目标并确定它们的类别与位置	用方框标注目标的四个顶点；工作量适中	Faster R-CNN^[65] YOLO v3^[66]
实例分割	确定目标个体的具体像素点	标注目标个体的边界像素点；工作量大	Mask R-CNN^[67] YOLACT^[68]
语义分割	像素级别的分类（不区分同类目标个体）	标注类别区域的边界像素点；工作量大	U-Net^[69] SegNet^[70]

视觉任务	功能	数据标签	代表算法
图像分类	将图像分到预定义类别中的某一类	说明图像所属的类别；工作量少	ResNet^[63] Vision Transformer^[64]
目标检测	识别所有目标并确定它们的类别与位置	用方框标注目标的四个顶点；工作量适中	Faster R-CNN^[65] YOLO v3^[66]
实例分割	确定目标个体的具体像素点	标注目标个体的边界像素点；工作量大	Mask R-CNN^[67] YOLACT^[68]
语义分割	像素级别的分类（不区分同类目标个体）	标注类别区域的边界像素点；工作量大	U-Net^[69] SegNet^[70]

内容	研究方向	瓶颈问题
原位成像	复杂结晶环境成像；高时空分辨率成像	复杂环境下的高质量成像；光学显微镜的分辨率限制
图像分析方法	基于图像大数据的结晶机理研究；更高精度的图像分析算法；图像标注需求较低的算法；轻量级算法；与其他PAT技术的协同分析	图像分析模型的泛化性能；兼顾模型的高精度与实时性；兼顾模型的高精度与数据标注工作量；多模态数据融合策略
结晶控制	基于图像的在线闭环控制；多模态协同控制	准确的结晶过程模型；兼顾控制策略的实时性与精度
三维成像	基于三维成像的结晶机理研究；高分辨率三维成像；高浓度悬浮液成像的三维重构；复杂工业环境的三维成像设备；复杂高分辨率重构的实时性	通用性强的三维图像重构算法；动态过程的三维实时追踪算法
其他领域	其他多相流类型的成像研究，包括气固、液液、气液等	图像分析模型的适用性与高精度

内容	研究方向	瓶颈问题
原位成像	复杂结晶环境成像；高时空分辨率成像	复杂环境下的高质量成像；光学显微镜的分辨率限制
图像分析方法	基于图像大数据的结晶机理研究；更高精度的图像分析算法；图像标注需求较低的算法；轻量级算法；与其他PAT技术的协同分析	图像分析模型的泛化性能；兼顾模型的高精度与实时性；兼顾模型的高精度与数据标注工作量；多模态数据融合策略
结晶控制	基于图像的在线闭环控制；多模态协同控制	准确的结晶过程模型；兼顾控制策略的实时性与精度
三维成像	基于三维成像的结晶机理研究；高分辨率三维成像；高浓度悬浮液成像的三维重构；复杂工业环境的三维成像设备；复杂高分辨率重构的实时性	通用性强的三维图像重构算法；动态过程的三维实时追踪算法
其他领域	其他多相流类型的成像研究，包括气固、液液、气液等	图像分析模型的适用性与高精度