Existing anomaly detection (AD) approaches rely on various hand‐crafted representations to represent video data and can be costly. The choice or designing of hand‐crafted representation can be difficult when faced with a new dataset without prior knowledge. Motivated by feature learning, e.g. deep leaning and the ability to directly learn useful representations and model high‐level abstraction from raw data, the authors investigate the possibility of using a universal approach. The objective is learning data‐driven high‐level representation for the task of video AD without relying on hand‐crafted representation. A deep incremental slow feature analysis (D‐IncSFA) network is constructed and applied to directly learning progressively abstract and global high‐level representations from raw data sequence. The D‐IncSFA network has the functionalities of both feature extractor and anomaly detector that make AD completion in one step. The proposed approach can precisely detect global anomaly such as crowd panic. To detect local anomaly, a set of anomaly maps, produced from the network at different scales, is used. The proposed approach is universal and convenient, working well in different types of scenarios with little human intervention and low memory and computational requirements. The advantages are validated by conducting extensive experiments on different challenge datasets.
Purpose
This paper aims to improve the generalization capability of feature extraction scheme by introducing a micro-cracks detection method based on self-learning features. Micro-cracks detection of multicrystalline solar cell surface based on machine vision is fast, economical, intelligent and easier for on-line detection. However, the generalization capability of feature extraction scheme adopted by existed methods is limited, which has become an obstacle for further improving the detection accuracy.
Design/methodology/approach
A novel micro-cracks detection method based on self-learning features and low-rank matrix recovery is proposed in this paper. First, the input image is preprocessed to suppress the noises and remove the busbars and fingers. Second, a self-learning feature extraction scheme in which the feature extraction templates are changed along with the input image is introduced. Third, the low-rank matrix recovery is applied to the decomposition of self-learning feature matrix for obtaining the preliminary detection result. Fourth, the preliminary detection result is optimized by incorporating the superpixel segmentation. Finally, the optimized result is further fine-tuned by morphological postprocessing.
Findings
Comprehensive evaluations are implemented on a data set which includes 120 testing images and corresponding human-annotated ground truth. Specifically, subjective evaluations show that the shape of detected micro-cracks is similar to the ground truth, and objective evaluations demonstrate that the proposed method has a high detection accuracy.
Originality/value
First, a self-learning feature extraction method which has good generalization capability is proposed. Second, the low-rank matrix recovery is combined with superpixel segmentation for locating the defective regions.
In visual tracking, holistic and part-based representations are both popular choices to model target appearance. The former is known for great efficiency and convenience while the latter for robustness against local appearance or shape variations. Based on non-negative matrix factorization (NMF), we propose a novel visual tracker that takes advantage of both groups. The idea is to model the target appearance by a non-negative combination of non-negative components learned from examples observed in previous frames. To adjust NMF to the tracking context, we include sparsity and smoothness constraints in addition to the non-negativity one. Furthermore, an online iterative learning algorithm, together with a proof of convergence, is proposed for efficient model updating. Putting these ingredients together with a particle filter framework, the proposed tracker, Constrained Online Non-negative Matrix Factorization (CONMF), achieves robustness to challenging appearance variations and non-trivial deformations while runs in real time. We evaluate the proposed tracker on various benchmark sequences containing targets undergoing large variations in scale, pose or illumination. The robustness and efficiency of CONMF is validated in comparison with several state-of-the-art trackers.
We propose a novel local nearest neighbor distance (LNND) descriptor for anomaly detection in crowded scenes. Comparing with the commonly used low-level feature descriptors in previous works, LNND descriptor has two major advantages. First, LNND descriptor efficiently incorporates spatial and temporal contextual information around the video event that is important for detecting anomalous interaction among multiple events, while most existing feature descriptors only contain the information of single event. Second, LNND descriptor is a compact representation and its dimensionality is typically much lower than the low-level feature descriptor. Therefore, not only the computation time and storage requirement can be accordingly saved by using LNND descriptor for the anomaly detection method with offline training fashion, but also the negative aspects caused by using high-dimensional feature descriptor can be avoided. We validate the effectiveness of LNND descriptor by conducting extensive experiments on different benchmark datasets. Experimental results show the promising performance of LNND-based method against the state-of-the-art methods. It is worthwhile to notice that the LNND-based approach requires less intermediate processing steps without any subsequent processing such as smoothing but achieves comparable event better performance.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.