Hyperspectral Acquisition Technology Based on Compressed Sampling in Spatial Domain

In the hyperspectral imaging device, the sensor detects the reflection or radiation intensity of the target at hundreds of different wavelengths, thus forming a spectral image composed of hundreds of continuous bands. The traditional processing method of sampling first and then compressing not only cannot fundamentally solve the problem of huge amount of data, but also causes waste of resources. To solve this problem, a spectral image reconstruction method based on compressed sampling in spatial domain and transform coding in spectral domain is designed by using the sparsity of single-band two-dimensional image and the spectral redundancy of spatial coded data. Based on Bayesian theory, a compressed sensing measurement matrix of adaptive projection is proposed. Combining these two algorithms, an adaptive Grouplet-FBCS algorithm is constructed to reconstruct the image using smooth projection Landweber. Experimental results show that, compared with existing image block compression sensing algorithms, this algorithm can significantly improve the quality of image signal reconstruction.

A Visual Cryptography-Based Watermarking Approach for the Detection and Localization of Image Forgery

The usage of images in different fields has increased dramatically, especially in medical image analysis and social media. Many risks can threaten the integrity and confidentiality of digital images transmitted through the internet. As such, the preservation of the contents of these images is of the utmost importance for sensitive healthcare systems. In this paper, the researchers propose a block-based approach to protect the integrity of digital images by detecting and localizing forgeries. It employs a visual cryptography-based watermarking approach to provide the capabilities of forgery detection and localization. In this watermarking scheme, features and key and secret shares are generated. The feature share is constructed by extracting features from equal-sized blocks of the image by using a Walsh transform, a local binary pattern and a discrete wavelet transform. Then, the key share is generated randomly from each image block, and the secret share is constructed by applying the XOR operation between the watermark, feature share and key share. The CASIA V 1.0 and SIPI datasets were used to check the performance and robustness of the proposed method. The experimental results from these datasets revealed that the percentages of the precision, recall and F1 score classification indicators were approximately 97% for these indicators, while the percentages of the TAF and NC image quality indicators were approximately 97% and 96% after applying several known image processing and geometric attacks. Furthermore, the comparative experimental results with the state-of-art approaches proved the robustness and noticeable improvement in the proposed approach for the detection and localization of image forgeries in terms of classification and quality measures.

Parameter Estimation of Poisson–Gaussian Signal-Dependent Noise from Single Image of CMOS/CCD Image Sensor Using Local Binary Cyclic Jumping

Since signal-dependent noise in a local weak texture region of a noisy image is approximated as additive noise, the corresponding noise parameters can be estimated from a given set of weakly textured image blocks. As a result, the meticulous selection of weakly textured image blocks plays a decisive role to estimate the noise parameters accurately. The existing methods consider the finite directions of the texture of image blocks or directly use the average value of an image block to select the weakly textured image block, which can result in errors. To overcome the drawbacks of the existing methods, this paper proposes a novel noise parameter estimation method using local binary cyclic jumping to aid in the selection of these weakly textured image blocks. The texture intensity of the image block is first defined by the cumulative average of the LBCJ information in the eight neighborhoods around the pixel, and, subsequently, the threshold is set for selecting weakly textured image blocks through texture intensity distribution of the image blocks and inverse binomial cumulative function. The experimental results reveal that the proposed method outperforms the existing alternative algorithms by 23% and 22% for the evaluative measures of MSE (a) and MSE (b), respectively.

Projection-to-image transform frame: a lightweight block reconstruction network (LBRN) for computed tomography

To solve the problem of learning based computed tomography (CT) reconstruction, several reconstruction networks were invented. However, applying neural network to tomographic reconstruction still remains challenging due to unacceptable memory space requirement. In this study, we presents a novel lightweight block reconstruction network (LBRN), which transforms the reconstruction operator into a deep neural network by unrolling the filter back-projection (FBP) method. Specifically, the proposed network contains two main modules, which, respectively, correspond to the filter and back-projection of FBP method. The first module of LBRN decouples the relationship of Radon transform between the reconstructed image and the projection data. Therefore, the following module, block back-projection module, can use the block reconstruction strategy. Due to each image block is only connected with part filtered projection data, the network structure is greatly simplified and the parameters of the whole network is dramatically reduced. Moreover, this approach is trained end-to-end, working directly from raw projection data and does not depend on any initial images. Five reconstruction experiments are conducted to evaluate the performance of the proposed LBRN: full angle, low-dose CT, region of interest (ROI), metal artifacts reduction and real data experiment. The results of the experiments show that the LBRN can be effectively introduced into the reconstruction process and has outstanding advantages in terms of different reconstruction problems.

An Optimal Method for Extracting the Center of Light Stripes

A method for extracting the center of the light stripe to effectively reduce the environmental noise is proposed in this paper. The block matching algorithm is adapted to use the global information in the structured light image to group image blocks with similar light stripe structures. The center coordinates of the light stripe in each group of image blocks are extracted by the gray gravity method, and its average value is used as the final center of light stripes in the similar image block, which reduces the influence of random noise on the accuracy of the extraction algorithm.

Laser vision seam tracking system based on proximal policy optimization

Purpose This paper aims to propose a weld seam tracking method based on proximal policy optimization (PPO). Design/methodology/approach By constructing a neural network based on PPO and using the reference image block and the image block to be detected as the dual-channel input of the network, the method predicts the translation relation between the two images and corrects the location of feature points in the weld image. The localization accuracy estimation network (LAE-Net) is built to update the reference image block during the welding process, which is helpful to reduce the tracking error. Findings Off-line simulation results show that the proposed algorithm has strong robustness and performs well on the test set of curved seam images with strong noise. In the welding experiment, the movement of welding torch is stable, the molten material is uniform and smooth and the welding error is small, which can meet the requirements of industrial production. Originality/value The idea of image registration is applied to weld seam tracking, and the weld seam tracking network is built on the basis of PPO. In order to further improve the tracking accuracy, the LAE-Net is constructed and the reference images can be updated.

Intelligent Fault Diagnosis and Forecast of Time-Varying Bearing Based on Deep Learning VMD-DenseNet

Rolling bearings are important in rotating machinery and equipment. This research proposes variational mode decomposition (VMD)-DenseNet to diagnose faults in bearings. The research feature involves analyzing the Hilbert spectrum through VMD whereby the vibration signal is converted into an image. Healthy and various faults show different characteristics on the image, thus there is no need to select features. Coupled with the lightweight network, DenseNet, for image classification and prediction. DenseNet is used to build a model of motor fault diagnosis; its structure is simple, and the calculation speed is fast. The method of using DenseNet for image feature learning can perform feature extraction on each image block of the image, providing full play to the advantages of deep learning to obtain accurate results. This research method is verified by the data of the time-varying bearing experimental device at the University of Ottawa. Through the four links of signal acquisition, feature extraction, fault identification, and prediction, a mechanical intelligent fault diagnosis system has established the state of bearing. The experimental results show that the method can accurately identify four common motor faults, with a VMD-DenseNet prediction accuracy rate of 92%. It provides a more effective method for bearing fault diagnosis and has a wide range of application prospects in fault diagnosis engineering. In the future, online and timely diagnosis can be achieved for intelligent fault diagnosis.

Application of Deep Learning and Brain Images in Diagnosis of Alzheimer’s Patients

In the identification of which stages Alzheimer’s patients are in, the application of the medical imaging technology helps doctors give more accurate qualitative diagnoses. However, the existing research results are not effective enough in the acquisition of valuable information from medical images, nor can they make full use of other modal images that highlight different feature information. To this end, this paper studies the application of deep learning and brain images in the diagnosis of Alzheimer’s patients. First, the image preprocessing operations and the brain image registration process were explained in detail. Then, the image block generation process was given, and the degrees of membership to white matter, gray matter and cerebrospinal fluid were calculated, and the brain images were also preliminarily classified. Finally, a complete auxiliary diagnosis process for Alzheimer’s disease based on deep learning was provided, an improved sparse noise reduction auto-encoder network was constructed, and the brain image recognition and classification based on deep learning were completed. The experimental results verified the effectiveness of the constructed model.

Contour-Based Detection and Quantification of Tar Spot Stromata Using Red-Green-Blue (RGB) Imagery

Quantifying symptoms of tar spot of corn has been conducted through visual-based estimations of the proportion of leaf area covered by the pathogenic structures generated by Phyllachora maydis (stromata). However, this traditional approach is costly in terms of time and labor, as well as prone to human subjectivity. An objective and accurate method, which is also time and labor-efficient, is of an urgent need for tar spot surveillance and high-throughput disease phenotyping. Here, we present the use of contour-based detection of fungal stromata to quantify disease intensity using Red-Green-Blue (RGB) images of tar spot-infected corn leaves. Image blocks (n = 1,130) generated by uniform partitioning the RGB images of leaves, were analyzed for their number of stromata by two independent, experienced human raters using ImageJ (visual estimates) and the experimental stromata contour detection algorithm (SCDA; digital measurements). Stromata count for each image block was then categorized into five classes and tested for the agreement of human raters and SCDA using Cohen's weighted kappa coefficient (κ). Adequate agreements of stromata counts were observed for each of the human raters to SCDA (κ = 0.83) and between the two human raters (κ = 0.95). Moreover, the SCDA was able to recognize “true stromata,” but to a lesser extent than human raters (average median recall = 90.5%, precision = 89.7%, and Dice = 88.3%). Furthermore, we tracked tar spot development throughout six time points using SCDA and we obtained high agreement between area under the disease progress curve (AUDPC) shared by visual disease severity and SCDA. Our results indicate the potential utility of SCDA in quantifying stromata using RGB images, complementing the traditional human, visual-based disease severity estimations, and serve as a foundation in building an accurate, high-throughput pipeline for the scoring of tar spot symptoms.

Melanoma Recognition and Lesion Segmentation using Multi-Instance Learning

Melanoma is one of the deadliest forms of skin cancer, but early and accurate identification can significantly improve the survival rate of patients. In this paper, an end-to-end framework based on multi-instance learning is proposed for melanoma recognition and lesion segmentation simultaneously. To make full use from the information of high-resolution images, we take each image block (super-pixel) as an instance in a bag and use multi-instance learning based on a graph convolutional network to recognize melanoma. Moreover, skin lesion segmentation is derived from attention weights and is calibrated by classification probability vectors. As a result, the AUC of our method for melanoma recognition reaches 0.93, which is much higher compared with other related methods. Also, the Jaccard index (JA) of our method for melanoma-related skin lesion segmentation reaches 0.699. In our end-to-end approach, segmentation and recognition are treated as intimately coupled processes, and hence, a high JA is also an indication of the reliability of melanoma recognition. Collectively, these findings confirmed that our method effectively assists melanoma diagnosis.