Search

It is significant to identify the moldy status of stored maize by fungi infection in the early stage. Hyperspectral imaging (HSI) combined with the sparse auto-encoders (SAE) and convolutional neural network (CNN) algorithms was used to classify the moldy grades of maize kernels. The HSI data were obtained in the range of 400–1000 nm, and four grades from health to heavy mildew were distinguished using the measured fungal spores of maize. The depth spectral features were represented using SAE and the image features were extracted by CNN. K nearest neighbors, support vector machine (SVM), and partial least squares discriminant analysis classifiers were combined with the spectral and image features to establish classification models to identify the different moldy grades of maize kernels. The comparison results indicated that the fusion of SAE and CNN combined with the SVM classifier to construct the SAE-CNN-SVM model had the most satisfactory identification result with high correct recognition rates of 99.47% and 98.94% for the training and testing sets, respectively, and the values of sensitivity and specificity were 0.95–1. The moldy grades were presented intuitively on the maize image based on pixels or kernel-wise. Therefore, the HSI with the SAE-CNN-SVM model had good recognition ability for the early detection of moldy maize kernels, which could potentially provide technical support for the development of online detection of moldy maize kernels during storage. [ABSTRACT FROM AUTHOR]

Moving target detection is an important technique in visual surveillance systems. If a camera is freely moving, it becomes more difficult to detect a moving target, especially in the environment of a wide-range background. To compensate for the global motion of a wide-range background, a disparity-based adaptive multi-homography method is proposed. The proposed method comprises four steps: 1) feature point extraction; 2) generation of adaptive multi-homography matrices using motion grouping; 3) generation of background model and detection of background; and 4) target detection. Experimental results show that the proposed method can robustly detect moving targets in sequences taken by a freely moving camera. [ABSTRACT FROM PUBLISHER]

A noise suppression method for feature frequency extraction that is supplemented with multi-point data fusion was investigated in consideration of issues involving wind turbine vibration signals subject to high noise disturbance. The difficulty of extracting early weak fault features was examined as well. First, a de-noising and feature extraction method that uses EMD-Correlation was developed by adding empirical mode decomposition (EMD) and autocorrelation de-noising to the wavelet package transform under the effects of white noise and short-term disturbance noise in wind turbine vibration signals. Second, an EMD-Correlation analysis model for feature frequency extraction supplemented with multi-point data fusion was established with reference to adaptive resonance theory-2 to highlight the feature frequency of a possible early weak fault. Third, the results obtained with the actual and simulated fault vibration signals of wind turbine bearing faults and the outcomes of comparing the different feature frequency extraction methods show that the proposed method of EMD-Correlation that is supplemented with multi-point data fusion can not only effectively reduce white noise and short-term disturbance noise but can also extract the feature frequency of early weak faults. Finally, prototype hardware and software were developed for a wind turbine condition monitoring system based on the aforementioned fault feature extraction algorithms and tested in an actual wind turbine generation system. [ABSTRACT FROM AUTHOR]

Although ship detection methods based on deep learning have achieved remarkable progress, the design of the universal ship detection (USD) system is rarely studied. The main challenge of USD lies in the notorious domain bias and shift problem across multiple domains. This article implements USD based on domain-invariant representations to alleviate this issue. Specifically, the proposed method integrates a multilevel domain classification network (MDCN) and a domain-centric cut-paste module (DCM). First, the backbone network is facilitated to learn domain-independent image features from multiple domains through MDCN, thereby reducing the disturbance of domain-specific features to universal detector. Furthermore, the proposed method combines the domain-related synthetic samples generated by DCM to provide MDCN with strong supervision information, which further motivates the network to be more attentive to the domain-invariant representations at the instance level. Finally, we conduct experiments on multiple ship datasets in the synthetic aperture radar (SAR) and optical domains to verify the effectiveness of the method. The results show that the proposed method outperforms baseline by around 2.95% average precision (AP50), which achieves an effective USD system by complementing the information between domain-invariant representations. [ABSTRACT FROM AUTHOR]

• A lightweight and efficient BLS-based crack detection model is proposed to automatically classify and identify crack images. • The bottleneck structure of MobileNetV3 is introduced as the convolution module of the proposed broad learning system for feature extraction. • By training and validating with public and realistic crack image datasets, the proposed model achieves fast and efficient incremental learning as training samples increase. • The sliding window algorithm is introduced to efficiently recognize and locate cracks in the large-size image. Concrete surface crack detection is an important means to ensure infrastructure security, including bridges and tunnels. In recent years, computer vision has been widely adopted for concrete surface crack detection, and the intelligent crack detection method based on computer vision has achieved good results with the breakthrough of the deep learning algorithm. However, there are also some challenges, such as complex architecture, high computing burden, and time-consuming training. In this paper, a lightweight broad learning system detection method called MobileNetV3-BLS is proposed. First, the inverted residual structure with the linear bottleneck of MobileNetV3 is introduced as the convolution module of the proposed broad learning system for feature extraction. Then the extracted features are employed to randomly generate both mapping and enhancement nodes in a broad learning system, which are used to train the crack detection model. The proposed MobileNetV3-BLS can improve crack detection accuracy and training speed compared with other classical deep learning networks. On the other hand, the proposed method is a dynamic network of which the weights can be updated quickly to achieve incremental learning by adding new nodes and samples. [ABSTRACT FROM AUTHOR]

Convolutional neural networks (CNNs) have been widely used for synthetic aperture radar (SAR) target detection. Typical methods based on CNN have obtained favorable detection accuracy at the cost of high model complexity, and thus are difficult to be directly applied to real-time satellites on board as well as maritime rescue. To deal with this problem, this article proposes an efficient and lightweight target detection network incorporating soft quantization. First, to compensate for the lack of accuracy caused by lightweight networks, a feature fusion module called split bidirectional feature pyramid network is proposed to alleviate the interference of complex background on SAR images. Meanwhile, to adapt the lightweight network and the feature fusion module, a linear transformation module is presented to enhance the linear representation of the model via learnable parameters. Eventually, to make the model size smaller, a soft quantization algorithm is proposed to reduce the accuracy degradation caused by quantization errors. We validate the robustness of the model in several publicly available datasets. Experimental results show that our model achieves 97.0% detection accuracy on SAR ship detection dataset, with a 0.9% accuracy improvement compared to mainstream methods using less than $15\times $ the number of parameters and less than $6\times $ the number of flops. [ABSTRACT FROM AUTHOR]

Abstract: The singular points of fingerprints, namely core and delta, play an important role in fingerprint recognition and classification systems. Several traditional methods have been proposed; however, these methods cannot achieve the reliable and accurate detection of poor-quality fingerprints. In this paper, an algorithm is proposed which combines improved Poincaré index and multi-resolution analysis to detect singular points. Conventional Poincaré index method is improved on the basis of the Zero-pole Model analysis to detect singular points with different resolutions. A model is presented to extract the multi-resolution information of the fingerprint pattern; this model divides fingerprint image into nonoverlapping blocks corresponding to different block sizes on the basis of wavelet functions to compute multiple resolution directional fields, and block position shifting is performed on these resolution levels to capture the features of the ridge direction patterns, where the corresponding shifting intervals are based on Sampling theorem. The relationship of singularities detected by improved Poincaré index in different resolution directional fields is used to confirm singular points accurately and reliably. The combination of local and global information makes our algorithm more robust to noise than methods that use local information only, and the existence of this algorithm increases the insight into the nature of singular points extraction. The accuracy and reliability of the method are demonstrated by experiment on database NIST-4, public fingerprint databases FVC02 DB1 and DB2. [Copyright &y& Elsevier]