Image Rotation Research Articles

Recognition Insect Images at the Order Level Based on Elliptic Metric Learning

HighlightsInsect order recognition.Space transformation combined with elliptic metric.Feature fusion with the Relief-F for the sliding window feature and HOG feature.The recognition accuracy of insect images at the order level is 95.8%.Abstract. The accurate recognition of insects is crucial for the protection and management of crops, vegetables, and fruit trees. This study proposes a recognition algorithm for insect images at the order level based on elliptic metric learning to improve the accuracy of the recognition. The nonlinear transformation that reflects the space structure or semantic information of insect image features is determined via elliptic metric learning. Then, the model for the potential relationship among insect image features is built. This model reduces the distance between same-order features but increases the distance between different-order features, improving the discriminative performance of classifiers. In elliptic metric learning, the regular term of the Frobenius norm is added to the triple constraint function to avoid overfitting and improve generalization ability. Moreover, to weaken the effects of image scale, rotation, and other changes on the recognition result, the sliding window method is used to extract the HSV color feature and the scale-invariant local ternary pattern texture feature of insect images. The feature vector is formed by gathering the maximum value of the features in the same horizontal sliding window. The features of the histogram of oriented gradients are also extracted from the images. Then, the optimized features are gained using the Relief-F algorithm. The results of the experiment on insect images at nine order levels show that the proposed method can significantly improve the recognition accuracy of insect images at the order level. Keywords: Elliptic metric, Insect images at the order level, Object recognition, Triple constraint, Regular term.

A feature-based approach for digital camera identification using photo-response non-uniformity noise

Source camera identification of an image is an emerging field of digital forensics. To identify the source camera through which the image is captured, photo-response non-uniformity (PRNU) noise is used as a camera fingerprint, as it is a unique characteristic that distinguishes images taken from the similar cameras. This paper presents a feature-based approach to identify the source camera. The input image is denoised using the denoising filter and from this denoised image, PRNU noise pattern is extracted. These PRNU noise patterns are represented by Hu's invariants, which are perpetual under image scaling, translation and rotation. These features are fed to fuzzy min-max neural network (FMNN) for training and classification for digital camera identification. The proposed approach has the ability to identify the cameras capturing the same scene.

Deep learning SPECT lung perfusion image classification method based on attention mechanism

SPECT lung perfusion is an important functional imaging technology. It can capture the functional lesions of the lung in a non-invasive manner and has become an important clinical detection method for diseases such as pulmonary embolism. In order to realize the automatic detection of the degree of pulmonary embolism, this paper studies and constructs a deep classification model based on the attention mechanism. First, the normalization technique is used to convert the original lung perfusion file into a SPECT image; secondly, in view of the over-fitting phenomenon of the deep learning model caused by the small amount of medical image data and the unbalanced data, the image translation and rotation techniques are used to perform effective expansion; then, in order to improve the model’s feature extraction ability, the attention mechanism is combined with the depth classification model to build a SPECT lung perfusion image classification model; finally, a set of real SPECT lung perfusion images were used to carry out comparative experiments on various depth classification models. The experimental results show that the model proposed in this paper can effectively detect the extent of lung disease lesions, and the classification accuracy rate exceeds 88%, which verifies the effectiveness and reliability of the classification model.

Artifacts and quantitative biases in neutron tomography introduced by systematic and random errors

Systematic and random errors are often introduced in white-beam neutron Computed Tomography (CT) due to the nature of the neutron source, neutron-matter interactions or limitations in hardware. These errors can cause bias in measured linear attenuation coefficients (LACs) and artifacts in the reconstructed images, reducing image quality and impeding quantitative analysis. Three sources of error in neutron tomography were investigated and quantified that are particularly pertinent to spallation source facilities. These include detector to sample stage misalignment, beam fluctuations and undersampling due to missing CT projections. Quantitative analyses of the grayscale biases and misalignment artifacts were performed on the reconstructed data using ImageJ. Furthermore, the efficacy of classical iterative reconstruction algorithms for the suppression of missing projection artifacts was explored. Several image qualifying metrics were used to compare Filtered Back Projection (FBP) and iterative algorithms quantitatively. We show that beam intensity fluctuations if left uncorrected provide a minimal bias even when exaggerated across a stack of projections. In addition, detector-sample stage misalignment causes a geometrical misalignment blur when not corrected for and severe ring artifacts despite image rotation correction. We suggest that iterative algorithms are more favourable in suppressing missing projection neutron artifacts, however they inaccurately reproduce the LACs for missing-wedge reconstructions compared to FBP.

Two-flat crystal three-sided mutual test technology based on <i>N</i>-position image rotation method

光学平面面形的绝对检验技术规避了干涉仪参考面形精度的制约，能够有效提高平晶面形的检测精度。采用<i>N</i>次图像旋转法的两平晶三面互检的绝对检验技术，求解待测平晶的三维绝对面形分布，结果中包含了中频波段的信息。利用递推公式构造旋转变化项的<i>N</i>次虚拟旋转结果，求和平均后得到旋转变化项，叠加旋转不变项结果后得到待测波面面形。推导了算法的理论误差，针对旋转角度进行优化并增加虚拟旋转次数，提高了算法精度，优化后的仿真结果的残差波面的均方根值精度为0.14 nm。对150 mm口径平晶进行两平晶三面互检实验，并将实验结果与传统三面互检法结果进行比对，均方根值偏差小于0.5 nm，验证了算法的准确性。

Vision-Based Human Detection Techniques: A Descriptive Review

Cameras are being used everywhere for the safety and security of citizens in different countries. Using a machine to detect humans in a photo or a video frame is a very complicated and challenging task. Various techniques have been developed for this purpose, which mainly rely on Artificial Intelligence. This article aims to provide a comprehensive review and analysis of the literatures from a descriptive perspective, which is its main differentiator from the existing survey papers in this area. Firstly, the vision-based human detection techniques and classifiers are elucidated in conjunction with the variants of feature extraction techniques. Secondly, various pros and cons of such techniques are discussed. Then, an investigation has been conducted and reported based on the state-of-the-art human detection descriptors (e.g. Log-Average Miss Rate and accuracy). Although techniques such as Viola-Jones and Speeded-Up Robust Features can detect objects in real-time and overcome Scale-Invariant Feature Transform (SIFT) limitations, they are still sensitive to illuminated conditions. Other techniques such as SIFT, Bag of Words, Orthogonal Moments, and Histogram of oriented Gradients provide other interesting benefits which include insensitivity to occlusion and clutters, simplicity, low-order element construction and invariance to illuminated conditions; nevertheless, they are computationally expensive and sensitive to image rotation. A meticulous review along similar lines revealed that the Deformable Part-based Model performs relatively better due to its ability to deal with particular pose variations and multiple views, occlusion handling (partial) and is application-free while its counterparts focus on only a single aspect. This article highlights and provides a brief description of each available data-sets for human detection research. Various use-cases of human detection systems are also elaborated. Finally, various conclusions are derived based on the conducted review followed by recommendations for future directions and possibilities to further improve the speed and accuracy of human detection systems.

Outside Box and Contactless Palm Vein Recognition Based on a Wavelet Denoising ResNet

Palm vein recognition is a high-security biometric. Outside the NIR capture box and contactless palm vein recognition are more popular but challenging. The users feel comfortable outside the NIR capture box but face more optical blurring brought by visible light. Contactless capture gestures solve the hygienic problem but face the image rotation, position translation, and scale transformation which makes classification difficult especially in large-scale databases. To address these problems, we develop a wavelet denoising ResNet, which consists of two models: the wavelet denoising (WD) model and the squeeze-and-excitation ResNet18 (SER) model. The WD model focuses on removing noise from skin scattering and optical blurring from palm vein images. The WD model enhances the low-frequency feature into a deep learning feature by residual learning technology. This strategy increases the weight of an effective handcrafted feature in the deep learning network. The SER model overcomes rotation, position translation, and scale transformation by selectively emphasizing classification features and weakening less useful features. To train and verify the network, an inside box palm vein image database and an outside box palm vein image database are set up. The Tongji contactless palm vein image database was also employed in the experiments. The validity and superiority of our network are verified in a series of experiments.

Deep learning based automatic segmentation of metastasis hotspots in thorax bone SPECT images.

SPECT imaging has been identified as an effective medical modality for diagnosis, treatment, evaluation and prevention of a range of serious diseases and medical conditions. Bone SPECT scan has the potential to provide more accurate assessment of disease stage and severity. Segmenting hotspot in bone SPECT images plays a crucial role to calculate metrics like tumor uptake and metabolic tumor burden. Deep learning techniques especially the convolutional neural networks have been widely exploited for reliable segmentation of hotspots or lesions, organs and tissues in the traditional structural medical images (i.e., CT and MRI) due to their ability of automatically learning the features from images in an optimal way. In order to segment hotspots in bone SPECT images for automatic assessment of metastasis, in this work, we develop several deep learning based segmentation models. Specifically, each original whole-body bone SPECT image is processed to extract the thorax area, followed by image mirror, translation and rotation operations, which augments the original dataset. We then build segmentation models based on two commonly-used famous deep networks including U-Net and Mask R-CNN by fine-tuning their structures. Experimental evaluation conducted on a group of real-world bone SEPCT images reveals that the built segmentation models are workable on identifying and segmenting hotspots of metastasis in bone SEPCT images, achieving a value of 0.9920, 0.7721, 0.6788 and 0.6103 for PA (accuracy), CPA (precision), Rec (recall) and IoU, respectively. Finally, we conclude that the deep learning technology have the huge potential to identify and segment hotspots in bone SPECT images.

Rotational invariant biologically inspired object recognition

The biologically inspired hierarchical model and X (HMAX) has been one of the superior techniques for object recognition purposes. HMAX is a robust method in the presence of some image variations including illumination, different scales, and location changes. However, the performance of HMAX extremely deteriorates if the orientation of the applied images in training phase is different than the orientation of testing images. In this study, the authors propose rotational invariant HMAX (RIMAX) to overcome the existing issues in object recognition imposed by rotations in the images. To this end, they embed two new layers into the structure of the standard HMAX. In the first added layer, non-accidental properties (e.g. corners and edges) are extracted as features that lead to obtaining a repeatable object recognition process. The second added layer provides robustness of RIMAX to image rotations by normalising the dominant orientation of the extracted features. Furthermore, they considerably reduce the imposed computational load by modifying template matching strategy as well as removing multiple scales of the Gabor filter. Simulation results employ Caltech101, TUD, Caltech5, and GRAZ-02 databases that validate RIMAX outperforms the standard HMAX and the other mathematical approaches in terms of robustness, accuracy, repeatability, and speed.

Detection of Bacterial Wilt on Enset Crop Using Deep Learning Approach

Bacterial Wilt disease is the most determinant factor as it results in a serious reduction in the quality and quantity of food produced by Enset crop. Therefore, early detection of Bacterial Wilt disease is important to diagnose and fight the disease. To this end, a deep learning approach that can detect the disease by using healthy and infected leave images of the crop is proposed. In particular, a convolutional neural network architecture is designed to classify the images collected from different farms as diseased or healthy. A total of 4896 images that were captured directly from the farm with the help of experts in the field of agriculture was used to train the proposed model. The proposed model was trained using these images and data augmentation techniques was applied to generate more images. Besides training the proposed model, a pre-trained model namely VGG16 is trained by using our dataset. The proposed model achieved a mean accuracy of 98.5% and the VGG16 pre-trained model achieved a mean accuracy of 96.6% by using a mini-batch size of 32 and a learning rate of 0.001. The preliminary results demonstrated that the effectiveness of the proposed approach under challenging conditions such as illumination, complex background, different resolutions, variable scale, rotation, and orientation of the real scene images.

Polarization state and image rotation via spontaneously generated coherence in a spinning fast light medium

In this article we propose a four-level rubidium (Rb87) atomic system for observing interesting features of polarization state rotation in a fast light medium. We investigate spontaneously generated coherence (SGC) for a spinning medium. We show how SGC can affect different spectral profiles of the polarization state and images in this suggested model. We observe a 0.5 radian rotation and 2.5 microsecond time advancement in our proposed system. Our precise results will provide a new platform for researchers in quantum optics due to its applications in image coding, telecommunications and cloaking technology.

Brain Age Prediction of Children Using Routine Brain MR Images via Deep Learning.

Predicting brain age of children accurately and quantitatively can give help in brain development analysis and brain disease diagnosis. Traditional methods to estimate brain age based on 3D magnetic resonance (MR), T1 weighted imaging (T1WI), and diffusion tensor imaging (DTI) need complex preprocessing and extra scanning time, decreasing clinical practice, especially in children. This research aims at proposing an end-to-end AI system based on deep learning to predict the brain age based on routine brain MR imaging. We spent over 5 years enrolling 220 stacked 2D routine clinical brain MR T1-weighted images of healthy children aged 0 to 5 years old and randomly divided those images into training data including 176 subjects and test data including 44 subjects. Data augmentation technology, which includes scaling, image rotation, translation, and gamma correction, was employed to extend the training data. A 10-layer 3D convolutional neural network (CNN) was designed for predicting the brain age of children and it achieved reliable and accurate results on test data with a mean absolute deviation (MAE) of 67.6 days, a root mean squared error (RMSE) of 96.1 days, a mean relative error (MRE) of 8.2%, a correlation coefficient (R) of 0.985, and a coefficient of determination (R2) of 0.971. Specially, the performance on predicting the age of children under 2 years old with a MAE of 28.9 days, a RMSE of 37.0 days, a MRE of 7.8%, a R of 0.983, and a R2 of 0.967 is much better than that over 2 with a MAE of 110.0 days, a RMSE of 133.5 days, a MRE of 8.2%, a R of 0.883, and a R2 of 0.780.

Multi-level 3D Densenets for False-positive Reduction in Lung Nodule Detection Based on Chest Computed Tomography.

False-positive nodule reduction is a crucial part of a computer-aided detection (CADe) system, which assists radiologists in accurate lung nodule detection. In this research, a novel scheme using multi-level 3D DenseNet framework is proposed to implement false-positive nodule reduction task. Multi-level 3D DenseNet models were extended to differentiate lung nodules from falsepositive nodules. First, different models were fed with 3D cubes with different sizes for encoding multi-level contextual information to meet the challenges of the large variations of lung nodules. In addition, image rotation and flipping were utilized to upsample positive samples which consisted of a positive sample set. Furthermore, the 3D DenseNets were designed to keep low-level information of nodules, as densely connected structures in DenseNet can reuse features of lung nodules and then boost feature propagation. Finally, the optimal weighted linear combination of all model scores obtained the best classification result in this research. The proposed method was evaluated with LUNA16 dataset which contained 888 thin-slice CT scans. The performance was validated via 10-fold cross-validation. Both the Free-response Receiver Operating Characteristic (FROC) curve and the Competition Performance Metric (CPM) score show that the proposed scheme can achieve a satisfactory detection performance in the falsepositive reduction track of the LUNA16 challenge. The result shows that the proposed scheme can be significant for false-positive nodule reduction task.

Digital Watermarks with Adaptive Information Ring Width in the Issue of Hidden Transmission of a Control Signal in the Multi-Target Robotic System

Purpose of research. The article is devoted to the issue of ensuring the safe transmission of control signals between the elements of a multi-agent robotic system. The purpose of the work is to provide hidden data transmission with the possibility of unambiguous extraction of control signals.Methods. To solve the task set, an algorithm based on the methods of digital watermarks and digital steganography was proposed. The method of forming a digital watermark in the form of a ring with symmetry allows not only ensuring maximum imperceptibility of embedding (insignificant distortion of the container during concealment), but also to protect the transmitted signal from such complex distortions as image rotation. The steganographic concealment method allows regulating the intensity of embedding using the force factor; it is computationally simple and straightforward. The proposed approach to identifying and understanding the transmitted signal differs from modern methods of cryptography and steganalysis in that it does not require 100% correct signal extraction. After several modifications, the method allows minimizing the time spent on formation (adaptive width of the ring with signal bits) and embedding of a digital watermark (minimization of the processed area of the container for embedding).Results. The proposed approach makes it possible to transmit control signals in a hidden way within the framework of the transmission of digital objects; the conducted experiments have shown that the control signal is unambiguously understood even with such distortions as a decrease or increase in contrast or brightness, image rotation, and compression.Conclusion. Application of the proposed technique for transmitting control signals in a multi-agent robotic system will male it possible to receive the necessary information timely and safe, with a minimum probability of loss.

Towards Better Surgical Instrument Segmentation in Endoscopic Vision: Multi-Angle Feature Aggregation and Contour Supervision

Accurate and real-time surgical instrument segmentation is important in the endoscopic vision of robot-assisted surgery, and significant challenges are posed by frequent instrument-tissue contacts and continuous change of observation perspective. For these challenging tasks more and more deep neural networks (DNN) models are designed in recent years. We are motivated to propose a general embeddable approach to improve these current DNN segmentation models without increasing the model parameter number. Firstly, observing the limited rotation-invariance performance of DNN, we proposed the Multi-Angle Feature Aggregation (MAFA) method, leveraging active image rotation to gain richer visual cues and make the prediction more robust to instrument orientation changes. Secondly, in the end-to-end training stage, the auxiliary contour supervision is utilized to guide the model to learn the boundary awareness, so that the contour shape of segmentation mask is more precise. The proposed method is validated with ablation experiments on the novel Sinus-Surgery datasets collected from surgeons' operations, and is compared to the existing methods on a public dataset collected with a da Vinci Xi Robot.

Image robust recognition based on feature-entropy-oriented differential fusion capsule network

In solving the black box attribute problem of neural networks, how to extract feature information in data and generalize inherent features of data are the focus of artificial intelligence research. Aiming at the problem of the weak generalization ability of large image transformation under deep convolutional networks, a new method for image robust recognition based on a feature-entropy-oriented differential fusion capsule network (DFC) is proposed, the core of which is feature entropy approximation. First, convolution feature entropy is introduced as the transformation metric at the feature extraction level, and a convolution difference scale space is constructed using a residual network to approximate the similar entropy. Then, based on this scale feature, convolution feature extraction in a lower scale space is carried out and fused with the last scale feature to form a convolution differential fusion feature. Finally, a capsule network is used to autonomously cluster using dynamic routing to complete the semantic learning of various high-dimensional features, thereby further enhancing the recognition robustness. Experimental results show that feature entropy can effectively evaluate the transformation image recognition effect, and the DFC is effective for robust recognition with large image transformations such as image translation, rotation, and scale transformation.

Clinically Relevant Vulnerabilities of Deep Machine Learning Systems for Skin Cancer Diagnosis

Clinically Relevant Vulnerabilities of Deep Machine Learning Systems for Skin Cancer Diagnosis

A Robust Forgery Detection Method for Copy–Move and Splicing Attacks in Images

Internet of Things (IoT) image sensors, social media, and smartphones generate huge volumes of digital images every day. Easy availability and usability of photo editing tools have made forgery attacks, primarily splicing and copy–move attacks, effortless, causing cybercrimes to be on the rise. While several models have been proposed in the literature for detecting these attacks, the robustness of those models has not been investigated when (i) a low number of tampered images are available for model building or (ii) images from IoT sensors are distorted due to image rotation or scaling caused by unwanted or unexpected changes in sensors’ physical set-up. Moreover, further improvement in detection accuracy is needed for real-word security management systems. To address these limitations, in this paper, an innovative image forgery detection method has been proposed based on Discrete Cosine Transformation (DCT) and Local Binary Pattern (LBP) and a new feature extraction method using the mean operator. First, images are divided into non-overlapping fixed size blocks and 2D block DCT is applied to capture changes due to image forgery. Then LBP is applied to the magnitude of the DCT array to enhance forgery artifacts. Finally, the mean value of a particular cell across all LBP blocks is computed, which yields a fixed number of features and presents a more computationally efficient method. Using Support Vector Machine (SVM), the proposed method has been extensively tested on four well known publicly available gray scale and color image forgery datasets, and additionally on an IoT based image forgery dataset that we built. Experimental results reveal the superiority of our proposed method over recent state-of-the-art methods in terms of widely used performance metrics and computational time and demonstrate robustness against low availability of forged training samples.

Prediction of Gleason Grade Group of Prostate Cancer on Multiparametric MRI using Deep Machine Learning Models

The Gleason Grade (GG) Group system has been introduced recently for more accurate stratification of prostate cancer (PCa). The grading system is based on the histologic patterns which is accessed from needle core biopsy, therefore it could be negatively impacted by the intratumor heterogeneity. We aim to develop a deep learning algorithm to predict GG groups using multiparametric magnetic resonance images (mp-MRI). We studied a retrospective collection of 201 patients with 320 lesions from the SPIE-AAPM-NCI PROSTATEx Challenge (https://doi.org/10.7937/K9TCIA.2017.MURS5CL), among which 98 patients with 110 lesions with GG available from biopsy. And the number of lesions in each subgroup was 36, 39, 20, 8, and 7, respectively, for GG 1-5. The images were acquired on two different types of Siemens 3T MR scanners. T2W images were acquired using a turbo spin echo sequence and had a resolution of around 0.5 mm in plane and a slice thickness of 3.6 mm. The DWI series were acquired with a single-shot echo planar imaging sequence with a resolution of 2 mm in-plane and 3.6 mm slice thickness and with diffusion-encoding gradients in three directions. Three b-values were acquired (50, 400, and 800 s/mm2), and subsequently, the ADC map was calculated by the scanner software. Image pre-processing included registration and normalization. Image rotation and scaling were also used to increase the sample size and re-balance the number of lesions in various GG. To prevent over-fitting on a small sample size, we implemented a transfer learning model by carrying over the features learned from the malignancy classification of 320 lesions from our previous model into the GG prediction. And we replaced the end-to-end convolutional neural network (CNN) training model with a combination of feature extraction using CNN and classification using weighted extreme learning machine (wELM). Features from the best performing model were extracted to represent each lesion, and those from the last convolutional layer were found constantly better than from all other layers. Based on 3-fold cross validation, the average validation results for sensitivity, specificity, positive predictive value, and negative predictive value for differentiation of each GG (1-5) were (1, 0.99, 0.97, 1), (0.69, 0.85, 0.73, 0.83), (0.9, 0.69, 0.46, 0.97), (0.89, 0.64, 0.16, 0.99), and (1, 0.78, 0.39, 1), respectively. GG4 had the highest false positive values. GG 3 was often misclassified as GG 4. Results of GG3-5 vs. GG1-2 were (0.82, 0.87, 0.76, 0.92). The stratification of GG4-5 vs. GG1-3 was (0.87, 0.81, 0.42, 0.98). This work has made substantial progress tackling the challenging task of GG prediction from mp-MRI due to a smaller and unbalanced data size by transferring knowledge from a malignancy classification task we developed earlier. The combined feature extraction using deep learning model and weighted extreme learning machine classifier has shown promising results for the GG prediction. This work was supported by a Research Scholar Grant, RSG-15-137-01-CCE from the American Cancer Society.

TGSIFT: Robust SIFT Descriptor Based on Tensor Gradient for Hyperspectral Images

This paper proposes a robust sparse descriptor based on tensor theory by using the spatial and spectral information synthetically, namely the Tensor gradient SIFT (TGSIFT), for Hyperspectral image (HSI). TGSIFT integrates both spatial and spectral information and considers the natural vector feature of HSIs. Based on the HSI Gaussian scale space, a new tensor model for HSI is proposed which takes the vectorial nature of HSI into consideration and preserves all the necessary structural information distributed over all the bands. The TGSIFT descriptor is formed based on the model proposed. Experimental results of HSI matching show that the TGSIFT descriptor achieves better matching performance than other SIFT descriptors under different transformations, including illumination change, sensor noise, image rotation, viewpoint change, and scale change.