浏览全部资源
扫码关注微信
Received：11 November 2021，
Revised：2022-6-28，
Accepted：05 July 2022，
Published：16 October 2022
移动端阅览
恶劣复杂及对抗环境下产生的图像数据通常受限，常具有非完整、不确定、小样本、小目标的全部或部分特性，与通常的图像大数据相比，对受限图像数据处理和分析的方法有明显区别：大数据的统计特性显著依赖于中心极限定理下的3σ原则，而受限下的图像数据统计一致性弱，难以体现可信且鲁棒的集中优势特点；遮挡、伪装等情况导致样本信息乃至维度都具有不完整性或非确定性，以模糊数学为基础的系列处理方法导致计算量剧增；以深度学习为代表的系列大数据处理方法得到巨大发展，但由于受限图像数据的处理和分析基本属于不可逆的逆问题，其解空间一般为高维子空间，如何可信鲁棒地确定空间中的解，目前尚无有效可行的通用理论和方法；采用隶属度为测度的系列方法需紧密依赖融入先验知识构造的代价函数。为促进受限图像数据的研究，梳理了对其处理和分析的机理、方法、手段以及遇到的困难，提出了可能的突破方法，以及催生的研究范式改变，以求抛砖引玉，吸引更多学者从事该研究。
Image big data and easy-use models such as deep networks(DNs) have expedited artificial intelligence (AI) currently. But
the issue for limited images which are often captured from complex
hostile and other scenarios has been challenging still yet: objects are too small to be recognized; the boundaries are fuzzy-overlapped; or even all the object information indicated in the images is uncertain due to camouflages and occlusions. The limited image data are featured with small-sample
small-object
incompleteness
uncertainty. Obviously
the tackle of limited image data is different from image big data: 1) image big data fits Gaussian distributions (mean
u
and σ) in terms of statistical central limit theorem
especially when data scales are much larger than data dimensions. This feature is beneficial to make statistical inference by the 3σ rule
which indicates that 99.73% of samples are within the range of [
u
-3σ
u
+3σ]. Possibly due to the concentration saliency fundamental
the statistical AI models such as DNs become very popular and seemly successful. However
for small dataset
the statistical consistency is usually poor
and the robust features cannot be identified based on concentration saliency. Thus
the statistical inference AI models are not feasible for small training data. 2) Image objects are often mutually occluded in costly and rare scenarios
and the delicate camouflages and masks
poor and complex luminous environment
as well as hostile disturbances
often make the image information itself or the information dimensions incomplete and unconfident. These issues make the computation load very heavy because the missed information or dimension has a large number of possibilities. 3) A large number of big data techniques have been proposed and seem very competitive in image processing field. For example
the DNs have achieved the best rank in many image big datasets publicly available. These features have extremely highlighted the contribution of DNs. However
many face recognition systems do not meet the requirement of accuracy and precision even when the face data is big enough. The statistical inference models such DNs cannot do precise inference
and some errors do exist due to the statistical inference itself. For limited image data
the consequence has been declined further especially when the number of samples is less than the dimensions. The partition boundaries in the sample space cannot be uniquely fixed by training samples. That is to say
to determine the inference model needs to solve irreversible inverse problems
meaning that the definite model cannot be uniquely fixed
and only a reduced model can be fixed. For any query sample
the simplified model cannot give an explicit solution
and only a subspace can just be given which is constituted of possible solutions. How to choose an appropriate solution from the subspace is still challenging
and seemingly there are no effective and general ways. Many techniques seem a little effective on studying limited image dataset
such as level-set methods
fussy logic methods
all these methods are based on the probability metrics measuring the divergence between the existent limited image data and the apriori knowledge or the specific background. That is to say
the cost functions indicating membership degrees
levels etc. are very critical for these methods.
Abdar M, Pourpanah F, Hussain S, Rezazadegan D, Liu L, Ghavamzadeh M, Fieguth P, Cao X C, Khosravi A, Acharya U R, Makarenkov V and Nahavandi S. 2021. A review of uncertainty quantification in deep learning: techniques, applications and challenges. Information Fusion, 76: 243-297[DOI: 10.1016/j.inffus.2021.05.008]
Altmann Y, McLaughlin S, Padgett M J, Goyal V K, Hero A O and Faccio D. 2018. Quantum-inspired computational imaging. Science, 361(6403): #2298[DOI: 10.1126/science.aat2298]
Chadès I and Nicol S. 2016. Small data call for big ideas. Nature, 539(7627): #31[DOI: 10.1038/539031e]
Davari A, Özkan H C, Maier A and Riess C. 2019. Fast and efficient limited data hyperspectral remote sensing image classification via GMM-based synthetic samples. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 12(7): 2107-2120[DOI: 10.1109/JSTARS.2019.2916495]
Dempster A P, Laird N M and Rubin D B. 1977. Maximum likelihood from incomplete data via the EM algorithm. Journal of the Royal Statistical Society: Series B (Methodological), 39(1): 1-22[DOI: 10.1111/j.2517-6161.1977.tb01600.x]
Goodfellow I J, Pouget-Abadie J, Mirza M, Xu B, Warde-Farley D, Ozair S, Courville A and Bengio Y. 2014. Generative adversarial nets//Proceedings of the 27th International Conference on Neural Information Processing Systems. Montreal, Canada: MIT Press: 2672-2680
Gu C Z, Lu X Q, He Y and Zhang C. 2020. Blur removal via blurred-noisy image pair. IEEE Transactions on Image Processing, 30: 345-359[DOI: 10.1109/TIP.2020.3036745]
Gull S F and Daniell G J. 1978. Image reconstruction from incomplete and noisy data. Nature, 272(5655): 686-690[DOI: 10.1038/272686a0]
Hansen P C. 1998. Rank-Deficient and Discrete ill-posed problems. Philadelphia, USA: SIAM[DOI: 10.1137/1.9780898719697]
He X, Chen Y S and Ghamisi P. 2021. Dual graph convolutional network for hyperspectral image classification with limited training samples. IEEE Transactions on Geoscience and Remote Sensing, 60: #5502418[DOI: 10.1109/TGRS.2021.3061088]
Hwang B, Woo T and Park J H. 2019. Fast diffraction-limited image recovery through turbulence via subsampled bispectrum analysis. Optics Letters, 44(24): 5985-5988[DOI: 10.1364/OL.44.005985]
Jayavadivel R and Prabaharan P. 2021. Investigation on automated surveillance monitoring for human identification and recognition using face and iris biometric. Journal of Ambient Intelligence and Humanized Computing, 12(11): 10197-10208[DOI: 10.1007/s12652-020-02787-1]
Jingade R R and Kunte R S. 2022. DOG-ADTCP: a new feature descriptor for protection of face identification system. Expert Systems with Applications, 201: #117207[DOI: 10.1016/j.eswa.2022.117207]
Kephart J O. 2011. Learning from nature. Science, 331(6018): 682-683[DOI: 10.1126/science.1201003]
Lan H R, Gong J L and Gao F. 2022. Deep learning adapted acceleration for limited-view photoacoustic image reconstruction. Optics Letters, 47(7): 1911-1914[DOI: 10.1364/OL.450860]
Li L Y, Cao G X, Liu J and Tong Y. 2021. Efficient detection in aerial images for resource-limited satellites. IEEE Geoscience and Remote Sensing Letters, 19: #6001605[DOI: 10.1109/LGRS.2020.3046739]
Liu Y Y, Zhao X L, Zheng Y B, Ma T H and Zhang H Y. 2021. Hyperspectral image restoration by tensor fibered rank constrained optimization and plug-and-playregularization. IEEE Transactions on Geoscience and Remote Sensing, 60: #5500717[DOI: 10.1109/TGRS.2020.3045169]
Mukherjee S and Awadallah A H. 2020. Uncertainty-aware self-training for few-shot text classification//Proceedings of the 34th International Conference on Neural Information Processing Systems. Vancouver, Canada: Curran Associates Inc. : #1780
Raudys S J and Jain A K. 1991. Small sample size effects in statistical pattern recognition: recommendations for practitioners. IEEE Transactions on Pattern Analysis and Machine Intelligence, 13(3): 252-264[DOI: 10.1109/34.75512]
Robbeloth M and Bourbakis N. 2016. Recognizing objects from their incomplete representation: a survey//Proceedings of the 7th International Conference on Information, Intelligence, Systems and Applications. Chalkidiki, Greece: IEEE: 1-9[ DOI:10.1109/IISA.2016.7785370 http://dx.doi.org/10.1109/IISA.2016.7785370 ]
Shafer G. 1976. A Mathematical Theory of Evidence. Princeton, USA: Princeton University Press
Trabelsi S, Samai D, Dornaika F, Benlamoudi A, Bensid K and Taleb-Ahmed A. 2022. Efficient palmprint biometric identification systems using deep learning and feature selection methods. Neural Computing and Applications, 34(14), 12119-12141[DOI: 10.1007/s00521-022-07098-4]
Vapnik V N. 2013. The Nature of Statistical Learning Theory. Berlin, Heidelberg, Germany: Springer
Wu P F, Liu Y G, Li Y Z and Liu B B. 2015. Robust prostate segmentation using intrinsic properties of TRUS images. IEEE Transactions on Medical Imaging, 34(6): 1321-1335[DOI: 10.1109/TMI.2015.2388699]
Yang L Q, Miao J F and Kou K I. 2022. Quaternion-based color image completion via logarithmic approximation. Information Sciences, 588: 82-105[DOI: 10.1016/j.ins.2021.12.055]
Yoo S, Yang X G, Wolfman M, Gursoy D and Katsaggelos A K. 2019. Sinogram image completion for limited angle tomography with generative adversarial networks//Proceedings of 2019 IEEE International Conference on Image Processing (ICIP). Taipei, China: IEEE: 1252-1256[ DOI:10.1109/ICIP.2019.8804416 http://dx.doi.org/10.1109/ICIP.2019.8804416 ]
Zhang Z Z, Lan C L, Zeng W J, Chen Z B and Chang S F. 2021. Uncertainty-aware few-shot image classification//Proceedings of the 30th International Joint Conference on Artificial Intelligence (IJCAI-21). Montreal, Canada: ijcai. org: 3420-3426[ DOI:10.24963/ijcai.2021/471 http://dx.doi.org/10.24963/ijcai.2021/471 ]
Zhou M Y, Chen H J, Paisley J, Ren L, Li L B, Xing Z M, Dunson D, Sapiro G and Carin L. 2012. Nonparametric Bayesian dictionary learning for analysis of noisy and incomplete images. IEEE Transactions on Image Processing, 21(1): 130-144[DOI:10.1109/TIP.2011.2160072]
相关文章
相关作者
相关机构
京公网安备11010802024621
