CA : A Cancer Journal for Clinicians

Texture features for image classification and retrieval.

The excitation and vibration triggered by the long-term operation of railway vehicles inevitably result in defective states of catenary support devices. With the massive construction of high-speed electrified railways, automatic defect detection of diverse and plentiful fasteners on the catenary support device is of great significance for operation safety and cost reduction. Nowadays, the catenary support devices are periodically captured by the cameras mounted on the inspection vehicles during the night, but the inspection still mostly relies on human visual interpretation. To reduce the human involvement, this paper proposes a novel vision-based method that applies the deep convolutional neural networks (DCNNs) in the defect detection of the fasteners. Our system cascades three DCNN-based detection stages in a coarse-to-fine manner, including two detectors to sequentially localize the cantilever joints and their fasteners and a classifier to diagnose the fasteners’ defects. Extensive experiments and comparisons of the defect detection of catenary support devices along the Wuhan–Guangzhou high-speed railway line indicate that the system can achieve a high detection rate with good adaptation and robustness in complex environments.

/pdf/automatic-defect-detection-of-fasteners-on-the-catenary-4p56mvzf63.pdf

Automatic Defect Detection of Fasteners on the Catenary Support Device Using Deep Convolutional Neural Network

Railroad tracks need to be periodically inspected and monitored to ensure safe transportation. Automated track inspection using computer vision and pattern recognition methods has recently shown the potential to improve safety by allowing for more frequent inspections while reducing human errors. Achieving full automation is still very challenging due to the number of different possible failure modes, as well as the broad range of image variations that can potentially trigger false alarms. In addition, the number of defective components is very small, so not many training examples are available for the machine to learn a robust anomaly detector. In this paper, we show that detection performance can be improved by combining multiple detectors within a multitask learning framework. We show that this approach results in improved accuracy for detecting defects on railway ties and fasteners.

Deep Multitask Learning for Railway Track Inspection

Railway track fastener defect detection based on image processing and deep learning techniques: A comparative study

The detection of fastener defects is an important task in railway inspection systems, and it is frequently performed to ensure the safety of train traffic. Traditional inspection is usually operated by trained workers who walk along railway lines to search for potential risks. However, the manual inspection is very slow, costly, and dangerous. This paper proposes an automatic visual inspection system for detecting partially worn and completely missing fasteners using probabilistic topic model. Specifically, our method is able to simultaneously model diverse types of fasteners with different orientations and illumination conditions using unlabeled data. To assess the damages, the test fasteners are compared with the trained models and automatically ranked into three levels based on the likelihood probability. The experimental results demonstrate the effectiveness of this method.

Automatic Fastener Classification and Defect Detection in Vision-Based Railway Inspection Systems

This paper has proposed an architecture of optimised SIFT (Scale Invariant Feature Transform) feature detection for an FPGA implementation of an image matcher. In order for SIFT based image matcher to be implemented on an FPGA efficiently, in terms of speed and hardware resource usage, the original SIFT algorithm has been significantly optimised in the following aspects: 1) Upsampling has been replaced with downsampling to save the interpolation operation. 2) Only four scales with two octaves are needed for our image matcher with moderate degradation of matching performance. 3) The total dimension of the feature descriptor has been reduced to 72 from 128 of the original SIFT, which leads to significantly simplify the image matching operation. With the optimisation above, the proposed FPGA implementation is able to detect the features of a typical image of 640×480 pixels within 31 milliseconds. Therefore, compared with the existing SIFT FPGA implementation, which requires 33 milliseconds for an image of 320×240 pixels, a significant improvement has been achieved for our proposed architecture.

An architecture of optimised SIFT feature detection for an FPGA implementation of an image matcher

Scale Invariant Feature Transform (SIFT) algorithm has the potential of detecting a large number of features from images, which makes the feature descriptor generation become a bottleneck of the processing speed and hence degrade the overall performance of the algorithm. To tackle this problem, we propose an efficient hardware architecture based on the polar sampled descriptor in this paper. It takes only 7.57us to generate a feature descriptor of 72 dimensions with a system frequency of 100MHz, which is equivalent to approximately 132100 feature descriptors per second. It can generate feature descriptors for VGA (640×480 pixels) resolution video at 60 frames per second (fps), provided that there are no more than 2200 features per frame. As far as we know, our hardware architecture has the highest processing speed for descriptor generation, compared with other existing architectures.

An efficient hardware architecture of the optimised SIFT descriptor generation

Bag-of-features (BoF) model based on SIFT generally assumes each descriptor is related to only one visual word of the codebook. Therefore, the potential correlation between the descriptor and other visual words is ignored. On the other hand, sparse coding through l1-norm regularization fails to generate optimal sparse representations since l1-norm regularization randomly selected one variable from a group of highly correlated variables. In this study we propose a novel bag-of-features model for image retrieval called SIFT-based Elastic sparse coding. The method utilizes a large number of SIFT descriptors to construct the codebook. The Elastic Net regression framework, which combines both l1-norm and l2-norm penalties, is then used to obtain the sparse-coefficient vector corresponding to the SIFT descriptor. Finally each image can be represented by a unified sparse-coefficient vector. Experimental results on Coil20 dataset demonstrate the consistent superiority of the proposed method over the state-of-the-art algorithms including original SIFT matching, conventional BoF strategy and BoF model based on l1-norm sparse coding.

SIFT-based Elastic sparse coding for image retrieval

Pathological image retrieval contributes to computer-aided diagnosis for breast cancer due to the fact that the retrieval results generally contain detailed diagnostic information (e.g. abnormal regions and diagnostic opinion from other doctors) which can offer some reference and assistance to the doctor during diagnosis process. In this paper, we present a novel pathological image retrieval approach based on probabilistic latent semantic analysis (pLSA) model. The method respectively utilizes SIFT features after visual saliency detection, and block Gabor features for the construction of two semantic codebooks, which not only can characterize the salient local invariant features and texture information under different scales and orientations in the pathological images, but also consider the high-level semantic features. Furthermore, we apply pLSA model to discover the latent topics in each codebook. Finally each pathological image is represented by the combination of topics from these two codebooks. The proposed method is evaluated on the pathological image database for breast cancer, which includes 5 categories (mucinous cystadenocarcinoma, invasive lobular carcinoma, basal-like carcinoma, invasive breast cancer and low-grade adenosquamous carcinoma) and 110 cases for each category. Experimental results demonstrate the feasibility and effectiveness of our method.

Hao Feng

Papers

Automatic Fastener Classification and Defect Detection in Vision-Based Railway Inspection Systems

An architecture of optimised SIFT feature detection for an FPGA implementation of an image matcher

An efficient hardware architecture of the optimised SIFT descriptor generation

SIFT-based Elastic sparse coding for image retrieval

Pathological Image Retrieval for Breast Cancer with pLSA Model