This study presents insights into the computational complexity of fractal image compression (FIC) algorithms. Unlike JPEG, a fractal encoder necessitates more CPU time in contrast to the decoder. The study examines various factors that impact the encoder and its computational cost. Many researchers have dedicated themselves to the field of fractal encoding to overcome the computational cost of the FIC algorithm. Here, this study offers a look over the approaches in the aspect of time complexity. The automated baseline fractal compression algorithm is studied to demonstrate the understanding of delay in the encoder. The study establishes how various approaches trade-off between the quality of decoder, compression ratio, and CPU time. The experiment section shows the bargain between fidelity criteria of the baseline algorithm.

Computer-aided diagnosis (CAD) is a common tool for the detection of diseases, particularly different types of cancers, based on medical images. Digital image processing thus plays a significant role in the processing and analysis of medical images for diseases identification and detection purposes. In this study, an efficient CAD system for the acute lymphoblastic leukaemia (ALL) detection is proposed. The proposed approach entails two phases. In the first phase, the white blood cells (WBCs) are segmented from the microscopic blood image. The second phase involves extracting important features, such as shape and texture features from the segmented cells. Eventually, on the extracted features, Naïve Bayes and k-nearest neighbour classifier techniques are implemented to identify the segmented cells into normal and abnormal cells. The performance of the proposed approach has been assessed through comprehensive experiments carried out on the well-known ALL-IDB data set of microscopic blood images. The experimental results demonstrate the superior performance of the proposed approach over the state-of-the-art in terms of accuracy rate in which achieved 98.7%.

The query image is usually a simple and single object in image retrieval, and the reference images in the database usually have many distractions. The precision of image retrieval can be greatly improved If the target regions in the database image are extracted during retrieval. So this paper proposes a Bow image retrieval method based on SSD target detection. First, the training gallery is manually annotated to record the location and size information. Second, the SSD target detection model is trained with the labeled training gallery to obtain the target object SSD model. Third, the SSD model is used to locate the similar target regions of the reference image and the query graph. Finally, the target region information is mapped into the convolutional features, and these feature vectors are used for image similarity matching. The performance of the proposed method is evaluated on Paris6k, Oxford5k, Paris106k and Oxford105k databases. The experimental results show that the accuracy of image retrieval will be greatly improved by adding optimization methods in the proposed image retrieval framework. The image retrieval accuracy of this method is higher than that of similar methods in recent years.

In this study, an efficient a deep learning architecture-based peripheral blood cell image recognition and classification is proposed using hybrid disruption-based salp-swarm and cat swarm (DSSCS)-based optimized convolutional neural networks (DSSCSCNNs) method. The DSSCSCNN method is employed to overcome the hyperparameter problem in CNN and it also helps this model to work on small peripheral blood cell data sets. In the DSSCSCNN method, the authors develop a binary coding technique that converts parameter tuning problems into an optimization problem. The original salp swarm algorithm is enhanced using a disruptive operator and salp swarm optimization algorithm to form the novel DSSCS algorithm which increases the diversity of the search space by providing higher classification accuracy. In this study, the CNNs use Vgg-16 architecture is used for training purposes. The global classification accuracy obtained when trained with the Vgg-16 model is 97%. This method establishes a fine-tuning process to develop a classifier trained using 15,976 images acquired from clinical practice. The proposed model gives improved performance in terms of accuracy, specificity, and sensitivity. In the WBC determination, the proposed approach has shown 100% achievement. It also provides the best overall classification accuracy of 99%.

Image segmentation is a key step in the process of image data processing. The quality of image segmentation will directly affect the accuracy of image cognitive understanding. The purpose of image segmentation is to divide the image into regions with specific semantics. For the simple linear iterative clustering (SLIC) algorithm, the feature equalisation parameters need to be set manually during image segmentation, which results in the lack of segmentation effects and slow processing time. By introducing the theory of intermediary mathematics, an improved adaptive SLIC super-pixel algorithm is proposed, which can adaptive generate characteristic equalisation parameters according to the specific situation of the image, thereby simplifying the operation steps and improving the image segmentation effect. After experimental verification and analysis, compared with the original SLIC algorithm and several other super-pixel contrast algorithms, the algorithm in this study can effectively shorten the processing time and achieve a better segmentation effect.

Currently, PCA (principal component analysis) is widely used in many neural networks and has become a crucial part of the convolutional neural network (CNN) feature extraction. However, whether PCA is suitable for this process remains to be elucidated. The authors proposed a new method called balanced principal component (BPC) that generates a balanced local feature and combines with CNN as a layer to cope with the fusion problem. Specifically, BPC layer includes regionalisation module and average compression PCA (AC-PCA) module. First, they used regionalisation module to generate some sub-region that focuses on the local feature in each view. Secondly, the AC-PCA module is a computational process that enlarges the feature matrix by PCA and eventually compacts the matrix to a one-dimensional (1D) vector by AC. Next, all 1D vectors are compacted by AC to obtain a multi-dimensional balance. Finally, they designed this layer with an end-to-end trainable structure to promote the feature extraction task of CNN. They addressed 3D shapes using a projection method that is pre-trained on ImageNet and migration learning on ModelNet dataset. By comparing with the state-of-the-art network, they achieved a significant gain in performance of retrieval and classification tasks.

Medical applications have a massive footprint in human's day-to-day life. Among that, MRI has a significant role, as it incorporates a significant impact on a brain tumour. Segmenting the tumour from MRI is substantial, but it is a time-consuming process. Both the normal and abnormal tissues found in the brain look similar, which increases the difficulty of the tumour detection process. The digital image needs to be processed to obtain an exact tumour detection result. The tumour detection process comprises five different stages, such as pre-processing, segmentation, feature extraction, feature selection, and classification. In this proposed work, hybrid wavelet Hadamard transform and grey-level co-occurrence matrix are included for feature extraction. Feature selection utilises sequential forward selection, which is an easy greedy search algorithm. This algorithm chooses only the predominant features for classification. The classification uses a hybrid support vector machine and adaptive emperor penguin optimisation. The experimental analysis shows the efficiency of the proposed work in terms of accuracy, specificity, and sensitivity values by computing the true positive, false positive, true negative, and false negative.

This study focuses on the detection and expulsion of noisy pixels from an image contaminated by impulsive noise. A noise detection approach is developed to avoid the misinterpretation of noise-free pixel as noisy. In order to design the noise removal algorithm, a probabilistic decision-based improved trimmed median filter (PDITMF) algorithm is proposed which is intended to work out the conflict related to the even number of noise-free pixels in the trimmed median filter. It deploys two new estimation techniques for de-noising, namely, improved trimmed median filter (ITMF) and patch else ITMF (PEITMF) as per noise density. At last, the noise detection approach is applied in the proposed PDITMF to build up a new technique called a probabilistic decision-based adaptive improved trimmed median filter (PDAITMF) algorithm. The proposed algorithms, PDITMF and PDAITMF experiment with many standard sample images. Simulation results show the proposed algorithms are capable of detecting and de-noising the contaminated image very efficiently and have a better visual representation. Under the authors' knowledge, the PDAITMF outperforms recently reported algorithms in context to peak signal-to-noise ratio as well as an image enhancement factor with the lower execution time at all noise densities.

The spectral amplitude of most natural images is approximately isotropic and follows the power law. In this study, the authors propose a new non-iterative blind image deconvolution algorithm that builds an isosceles curve model to approximate the spectrum amplitude of the real image. In the authors’ proposed algorithm, the optical transfer function (OTF) is obtained by comparing the reconstructed and degraded spectra. Then they employ the integrated multidirectional comprehensive estimation to reduce the OTF estimation error. The restored image is then obtained by applying the estimated OTF and the Wiener filter. Experiments on image deconvolution tasks indicate that the proposed algorithm provides a significant performance gain by obtaining an accurate OTF, reducing ringing artefacts compared with existing algorithms, and realising real-time image restoration.

Magnetic resonance imaging (MRI) is used to diagnose multiple sclerosis (MS) disease lesions in the brain. Diagnosis of MS disease from MRI images is an important and vital thing in today's world. This disease can cause many problems for people who have this disease and reduce the life expectancy in them. So, a strong approach is needed to overcome the challenges in this field. In this study, a method is presented based on convolutional neural networks to detect MS disease lesions from MRI. Four layers of convolution, two layers of pooling, three layers of ReLU are applied, and instead of a fully connected layer, a convolutional layer with a filter size of 1 × 1 has been used to reduce network parameters. Also, for network training, stochastic gradient descent with momentum has been used such that itgreatly improves the speed of learning. Convolutional neural network has a strong potential for MS disease diagnosis and provides good results without the need for lesions segmentation. Also, it has a low sensitivity to the challenges of blurring and different contrasts, and it shows a good performance. The proposed method in this study shows 99.66% accuracy, 99.98% sensitivity and 99.33% specificity.

In optical imaging systems, the aberration is an important factor that impedes realising diffraction-limited imaging. Accurate wavefront sensing and control play important role in modern high-resolution optical imaging systems nowadays. In this study, a simple model-based phase retrieval algorithm is proposed for accurate efficient wavefront sensing with high dynamic range. In the authors’ algorithm, a wavefront is represented by the Zernike polynomials, and the Zernike coefficients are solved by the least-squares-based non-linear optimisation method, i.e. the Lederberg–Marquardt algorithm, with multiple phase-diversity images. The numerical results show that the proposed algorithm is capable of retrieving wavefront with a large dynamic range up to seven wavelength and robust to noise. In comparison, the proposed algorithm is more efficient than the existing model-based technique and more accurate than existing Fourier - transformation-based iterative techniques.

The existing single image super-resolution methods based on deep learning cannot handle multiple degradations well, and the generated image tends to be blurred and over-smoothed due to poor generalisation ability. In this study, the authors propose a method based on a generative adversarial network (GAN) to deal with multiple degradations. In the generator network, blur kernel and noise level are used as input through dimensionality stretching strategy preprocessing to make full use of prior knowledge. In addition, three discriminators with different scales are used in the discriminator network to pay attention to the reconstruction of image details while focusing on the global consistency of the image. For the problems of vanishing gradient and mode collapse existing in GAN-based methods, a gradient penalty term is added in the loss function. Extensive experiments demonstrate that the proposed method not only can handle multiple degradations to obtain state-of-the-art performance, but also deliver visually credible results in real scenes.

Copy-move image forgery is one of the most popular image tampering technique which can be performed for vicious purposes. In this forgery technique, selected region is copied and pasted at different locations on the same image to produce a manipulated image. Such forgery is denigratory as it can alter the image content by hiding or appending visual information. In this study, the authors propose a novel keypoint-based technique to detect forged images sustaining composite attacks consisting of various combinations of geometrical and post-processing operations. In the authors' method, AKAZE and FAST techniques are used to extract keypoints from the image. Non-maximal value suppression with automatic contrast thresholding is performed during FAST keypoint extraction. SIFT and DAISY descriptors are computed corresponding to extracted keypoints. PCA is applied over SIFT and DAISY descriptors to discard lower components which are sensitive to distortions occurred in images. They apply a correlation-based nearest neighbour search technique to detect similarity among keypoint descriptors. HDBSCAN algorithm is applied to obtain matched keypoint clusters. Further, RANSAC algorithm is utilised for removal of keypoint outliers. In comparison to state-of-the-art techniques, their approach achieve high F-measure (%) and low FPR (%) for image-level as well as pixel-level copy-move forgery detection.

Superpixel segmentation approach, as a preprocessing step in computer vision tasks, groups pixels into perceptually coherence atomic regions to replace the pixel grid in images and reduce the primitives and redundancy of subsequent works. In this study, the authors proposed a fast and robust superpixel generation method based on non-iterative framework with the constraint of linear path. They collected neighbouring pixels as the initial superpixels one by one in the conventional order at first. To make the superpixels attach to the most object boundaries well and robust to noise, they defined a new distance measurement between pixels and superpixel seeds by considering the colour difference of pixels in the neighbourhood and along the linear path from the pixel to the seed. Meanwhile, they proposed a new way to set parameters adaptively based on the intrinsic quality of images. Then, they refined the initial superpixels by merging the smallest ones until the number of superpixel meets the expectation. The experimental results on clean and noisy images demonstrate that the proposed method is effective and presents a competitive performance in computational efficiency with the state-of-the-art real-time method.

A novel learning-based end-to-end network for stereo matching, named Multi-path Attention Stereo Matching (MPA-Net), is introduced in this study. Different from existing methods, the multi-path attention aggregation module is designed firstly, named MPA, which is a unified structure using three different parallel layers with a respective attention mechanism to extract the multi-scale informational features. Secondly, the method of cost volume construction, which differs from the traditional stereo matching methods, is extended. And then, the absolute difference between two input features is calculated. Furthermore, a u-shaped structure with 3D attention gate is selected as the encoder-decoder module. Specifically, the module is used to fuse the encoding features to their corresponding decoding features under the supervision of the authors' attention gate with skip-connection, and thus exploit more significant information for matching cost regularisation and disparity prediction. Finally, specific experiments are conducted to evaluate their network on SceneFlow, KITTI2012 and KITTI2015 data sets. The results show that their method achieves a better improvement in disparity maps prediction compared with some existing state-of-the-art methods on KITTI benchmark.

Pattern regeneration is one of the applications of reverse engineering technology in the textile field, which realises the process of textile-pattern regeneration-textile, and fundamentally provides an intelligent design means of textile. At present, the method of pattern regeneration in the jacquard fabric is to use the image segmentation algorithm to segment the image digitalised by unidirectional imaging, and then the segmented pattern could be identified to regeneration for the design of new fabrics. However, due to the concave and convex pattern textures on the surface of jacquard fabric, the traditional unidirectional imaging method cannot be used for the full characterisation of its structural information, resulting in unsatisfactory pattern segmentation effect. To solve this problem, a novel segmentation algorithm for jacquard patterns based on multi-view image fusion was proposed in this study. Based on multi-view image acquisition and fusion, the pattern image of jacquard fabric could be cluster-segmented by extracting the complete texture information of the fused image and the actual colour information of the calibrated image. Compared with the traditional unidirectional imaging method, the experimental results show that the enhanced texture information of the fused image is more workable for the pattern segmentation, it validates the effectiveness of the proposed method.

Deep learning techniques are recently being used in fundus image analysis and diabetic retinopathy detection. Microaneurysms are important indicators of diabetic retinopathy progression. The authors introduce a two-stage deep learning approach for microaneurysms segmentation using multiple scales of the input with selective sampling and embedding triplet loss. The proposed approach facilitates a region proposal fully convolutional neural network trained on segmented patches and a patch-wise refinement network for improving the results suggested by the first stage hypothesis. To enhance the discriminative power of the second stage refinement network, the authors use triplet embedding loss with a selective sampling routine that dynamically assigns sampling probabilities to the oversampled class patches. This approach introduces a relative improvement over the vanilla fully convolutional neural network on the Indian Diabetic Retinopathy Image Data set segmentation data set. The proposed segmentation is incorporated in a classification model to solve two downstream tasks for diabetic retinopathy detection and referable diabetic retinopathy detection. The classification tasks are trained on the Kaggle diabetic retinopathy challenge data set and evaluated on the Messidor data. The authors show that adding the segmentation enhances the classification performance and achieves comparable performance to the state-of-the-art models.

Frame interpolation is one of the most challenging tasks in the video processing field. Recent advances have demonstrated that the deep learning-based frame interpolation methods are promising. However, the experiments show that most existing deep learning-based methods have the same problem as traditional methods. When these algorithms handle severe occlusions, they will produce distortions, especially around the motion boundaries. To better synthesise the image of the motion areas, the authors design a mask-guided frame synthesis model, which consists of multiple components, based on deep convolutional neural networks. The proposed model first estimates the asymmetric bi-directional optical flows from the intermediate frame to the input frames. Then it estimates the occlusion-aware masks, which can compensate for the optical flow inaccuracy based on optical flows and correlation information. Finally, the warped frames are adaptively fused under the guidance of the masks to generate a high-quality intermediate frame. Furthermore, to generate more realistic video frames, they train the network model with the pixel-based loss and the feature-based loss in a step-by-step way. In the experiment, they analyse the proposed model and compare it with the high-performance methods, both qualitative and quantitative results show that their method performs better.

An algorithm is proposed for segmentation of digital images using the curve fitting method based on modified Vandermonde matrix and modified Gram-Schmidt method. Modified Vandermonde matrix is applied to linearly smoothed histogram data to find the coefficients of a polynomial of a given degree for fitting the data in a least square sense. Modified Gram-Schmidt method is applied to get polynomial coefficients to minimise the least square distance during the calculation. The threshold value is computed by locating the minimum number of pixels from the grey-level value. Threshold value is used with that of the input image to generate an output segmented image. The output segmented image is further improved by applying morphological operation which has made the segmented edge lines and regions free from any extraneous pixels. The final segmented image has been found to have sharp edges and filled in a region within the image contours. The experimental results show that the proposed method has produced superior performance compared to other state-of-art algorithms. Further, it is observed from the results that the proposed algorithm takes less program execution time than others in vogue algorithms.

Retinal vessel segmentation has important application value in clinical diagnosis. If experts manually segment the retinal vessels, the workload is heavy, and the result is strong subjectively. However, some existing automatic segmentation methods have the problems of incomplete vessel segmentation and low-segmentation accuracy. In order to solve the above problems, this study proposes a retinal vessel segmentation method based on task-driven generative adversarial network (GAN). In the generative model, a U-Net network is used to segment the retinal vessels. In the discriminative model, multi-scale discriminators with different receptive fields are used to guide the generative model to generate more details. On the other hand, in view of the uncontrollable characteristics of the data generated by the traditional GAN, a task-driven model based on perceptual loss is added to traditional GAN for feature matching, which makes the generated image more task-specific. Experimental results show that the accuracy, sensitivity, specificity and area under the receiver operating characteristic curve of the proposed method on data set digital retinal images for vessel extraction are 96.83, 80.66, 98.97 and 0.9830%, respectively.

It is extremely challenging to accomplish excellent accuracy for gesture recognition using an approach where complexity in computation time for recognition is less. This study compares accuracy in hand gesture recognition of a single viewpoint set-up with proposed two viewpoint set-up for different classification techniques. The efficacy of the presented approach is verified practically with various image processing, feature extraction and classification techniques. Two camera system make geometry learning and three-dimensional (3D) view feasible compared to a single camera system. Geometrical features from additional viewpoint contribute to 3D view estimation of the hand gesture. It also improves the classification accuracy. Experimental results demonstrate that the proposed method show escalation in recognition rate compared to the single-camera system, and also has great performance using simple classifiers like the nearest neighbour and decision tree. Classification within 1 s is considered as real-time in this study.

Recently, it is becoming a challenging work for person re-identification due to the problems of occlusion, blurring and posture. The key of effective person re-identification is to capture sufficient detailed features of a person's appearance in images. Different from previous methods, our method mainly focuses on fusing different visual clues only depending on the features of different levels and scales without additional assistance. The major contributions of our paper are the mixed pooling strategy with different kernels and the mixed loss function. Firstly, we adopt ResNet50 as our backbone. We have slightly modified the backbone, which does not use the down-sampling operation at the beginning of stage 4. Inspired by pyramid pooling structure, we pass the outputs of Res4 and Res5 through the average pooling layer and max pooling layer with different kernels and strides separately. Secondly, we combine the averaged triplet losses and the averaged softmax losses as the final loss of the whole network. Extensive experiments on three datasets (CUHK3, Market1501, DukeMTMC-reID) show that compared with many state-of-the-art methods in recent years, our model achieve higher accuracy.

In order to increase the detection effect of fuzzy vehicles and small size vehicles, an improved multitask cascaded convolutional neural network (IMC-CNN) based on mixed image enhancement is proposed. Firstly, contrast limited adaptive histogram equalisation and multi-scale Retinex are used to enhance images. Mixed image enhancement can effectively solve the problems of image blurring, low contrast and uneven illumination when the imaging environment is not ideal. IMC-CNN includes two stages: object location and object classification. The object location network based on multi-layer feature fusion can locate and extract the object from complex background, and output regions contain only a single vehicle object. The object classification network is a lightweight convolutional neural network with only two convolutional layers, which can effectively reduce information loss and improve the classification accuracy of small objects and fuzzy objects. In addition, online hard example mining algorithm and focal loss function are adopted in network training. These strategies can solve the problem of unbalance between positive and negative samples. To verify the validity of the proposed algorithm, the experiments are performed on SYIT-Fuzzy dataset and COCO-Vehicle dataset. Compared with Faster R-CNN, YOLO v4 and other recent models, the average classification accuracy of the proposed method is significantly increased.

Information hiding is a technique that conceals private information in a trustable carrier, making it imperceptible to unauthorised people. This technique has been used extensively for secure transmissions of multimedia, such as videos, animations, and images. This study proposes a novel Tetris-based data hiding scheme to flexibly hide more secret messages while ensuring message security. First, an L_Q L_Q square lattice Q is selected to determine the maximum embedding capacity, and then it is filled without gaps through rotating and sliding tetrominoes while making the shape of each tetromino different. Secondly, according to the decided Q, the reference matrix and corresponding look-up table are constructed and then used for secret messages embedding and extraction. In the authors approach, each pixel pair of the original image can be processed to conceal 4- or 6-bit secret messages. The experimental results show that their proposed Tetris-based scheme has excellent performance, exceeding the performance of some state-of-the-art schemes in both embedding capacity and visual quality. The proposed scheme also provides secure covert communication.

The water quality, contaminant migration characteristics, and emissions quantity of pollutants in the basin would have a great impact on aquatic creatures, agricultural irrigation, human life, and so on. In the aquaculture industry, because water colour can reflect the species and number of phytoplankton in the water, the water quality type can be obtained by analysing the colour of the aquaculture water using image processing techniques. Therefore, this study proposes an intelligent monitoring approach for water quality. The critical features of water colour images are extracted, and then using the machine learning methods, an intelligent system for water quality monitoring is established based on the fused random vector functional link network (RVFL) and group method of data handling (GMDH) model. The proposed approach presents a superior performance relative to other state-of-the-art methods, and it achieves an average predicting accuracy of 96.19% on the feature dataset. Experimental findings demonstrate the validity of the proposed approach, and it is accomplished efficiently for the monitoring of water quality.

In this work, the authors presented the indexing method for efficient retrieval of Kannada characters. Here the characters were segmented by the connected component method. Then features were extracted by scale invariant feature transform (SIFT) and histogram of oriented gradients (HOG) methods. These features were condensed by principal component analysis. They presented the indexing approach using K-dimensional tree (K-D tree) to improve the identification process. For the experiment, they used their own database. The results of the experiment show that the indexing prior is faster than conventional identification approaches in terms of time to script. From experimentation, they observe that fusion features achieve the maximum accuracy of 90% with varying principal component analysis features.

Hand held gun detection has an important application in both the field of video forensic and surveillance, because, gun is operative by hand only while committing any crime with it. The significant application encompasses the vulnerable places, such as around airport, marketplace, shopping malls, etc. In view of non-availability of relevant public data set, this study provides a newly created mimicked video data set for detection of gun carried by a person and entitled as Tripura University Video Data set for Crime-Scene-Analysis (TUVD-CSA). Effects of illumination, occlusion, rotation, pan, tilt, scaling of gun are effectively demonstrated in it. Moreover, the authors proposed an Iterative Model Generation Framework (IMGF) for gun detection, which is immune to scaling and rotation. Instead of locating the best matched object (gun) in the whole reference image to a query model via exhaustive search, IMGF searches only where the moving person carrying gun appears, which drastically reduces the computational overhead associated with a general template matching scheme. This has been employed by the background subtraction algorithm. Experimental results demonstrate that the proposed IMGF performs efficiently in gun detection with lesser number of true-negatives compared with the state-of-the-art methods.

Joint detection and tracking of multiple extended targets (ETs) from image observations is a challenging radar technology; especially for extended stealth targets (ESTs). This work provides a new approach for the ESTs tracking under the non-linear Gaussian system based on track-before-detect (TBD) approach. The sequential Monte Carlo cardinality-balanced multi-target multi-Bernoulli (SMC-CBMeMBer) filter provides a good framework to cope with TBD approach. However, this filter suffers from the particles’ degradation problem seriously; especially for ETs tracking. Recently, the cubature Kalman (CK)-CBMeMBer filter which employs a third-degree spherical-radical cubature rule has been proposed to handle the non-linear models, the CK-CBMeMBer filter is more accurate and more principled in mathematical terms compared to SMC-CBMeMBer filter. To this point, the authors address a TBD of ESTs with extended CK-CBMeMBer filter based on random matrix model (RMM), which is an efficient way to track ellipsoidal ESTs. In RMM-ESTs scenarios, although the extension ellipsoid is efficient, it may not be accurate enough because of lacking useful information, such as size, shape, and orientation. Therefore, they introduce a filter composed of sub-ellipses; each one is represented by a RMM. The results confirm the effectiveness and robustness of the proposed filter.

The combination of a small unmanned ground vehicle (UGV) and a large unmanned carrier vehicle allows more flexibility in real applications such as rescue in dangerous scenarios. The autonomous recovery system, which is used to guide the small UGV back to the carrier vehicle, is an essential component to achieve a seamless combination of the two vehicles. This study proposes a novel autonomous recovery framework with a low-cost monocular vision system to provide accurate positioning and attitude estimation of the UGV during navigation. First, the authors introduce a light-weight convolutional neural network called UGV-KPNet to detect the keypoints of the small UGV form the images captured by a monocular camera. UGV-KPNet is computationally efficient with a small number of parameters and provides pixel-level accurate keypoints detection results in real-time. Then, six degrees of freedom (6-DoF) pose is estimated using the detected keypoints to obtain positioning and attitude information of the UGV. Besides, they are the first to create a large-scale real-world keypoints data set of the UGV. The experimental results demonstrate that the proposed system achieves state-of-the-art performance in terms of both accuracy and speed on UGV keypoint detection, and can further boost the 6-DoF pose estimation for the UGV.

Crowd counting is getting more and more attention in our daily life, because it can effectively prevent some safety problems. However, due to scale variations and background noise in the image, such as buildings and trees, getting the accurate number from image is a hard work. In order to address these problems, this work introduces a new multi-scale supervised network. The proposed model uses part of vgg16 model as the backbone to extract feature. In the training process, a multi-scale dilated convolution module is added at the end of each stage of the backbone network to generate attention map with different resolutions to help the model focus on the head area in feature map. In addition, the dilated convolution adopts three dilation ratios to fit different sizes of head in the image. Finally, in order to get the high-quality density map with high-resolution, the authors employ the upsampling operation to restore the density map size to the quarter size of original image. A large number of experiments on these four datasets show that the proposed network has greatly improved the counting accuracy of many existing methods.

In recent years, consumer depth cameras have been widely used in digital entertainment and human-machine interaction due to the advantages of real-time performance and low cost. Facial depth maps have shown great potential in 3D-face-related studies. However, disadvantages of low resolution and precision limit its further applications. In this work, the authors propose an edge-guided convolutional neural network for single facial depth map super-resolution. It consists of two parts: an edge prediction sub-network and a depth reconstruction sub-network. The edge prediction sub-network generates an edge guidance map to guide the depth reconstruction sub-network to recover sharp edges and fine structures. Effective data augmentation methods are proposed as well. The network is patch-based and able to cope with any size of the input depth maps. In addition, it is insensitive to the face pose since the synthetic training dataset they generated covers a wide range of face poses. The proposed method is validated with three datasets including a synthetic facial depth data set, a real Kinect V2 facial depth data set and Middlebury Stereo Data set. Experimental results show that it outperforms the state-of-the-art methods on all the three data sets.

With the development of artificial intelligence and image processing technology, more and more intelligent diagnosis technologies are used in cervical cancer screening. Among them, the detection of cervical lesions by thin liquid-based cytology is the most common method for cervical cancer screening. At present, most cervical cancer detection algorithms use the object detection technology of natural images, and often only minor modifications are made while ignoring the specificity of the complex application scenario of cervical lesions detection in cervical smear images. In this study, the authors combine the domain knowledge of cervical cancer detection and the characteristics of pathological cells to design a network and propose a booster for cervical cancer detection (CCDB). The booster mainly consists of two components: the refinement module and the spatial-aware module. The characteristics of cancer cells are fully considered in the booster, and the booster is light and transplantable. As far as the authors know, they are the first to design a CCDB according to the characteristics of cervical cancer cells. Compared with baseline (Retinanet), the sensitivity at four false positives per image and average precision of the proposed method are improved by 2.79 and 7.2%, respectively.

Image mosaicking is a series of successive algorithms that use a sequence of images or a video of a scene to create a single image with a wider field of view of the whole scene. In most cases, when dynamic objects exist in the input data, issues, such as ghosting, parallax effects or object duplication, are visible in the resulting mosaic. These technical errors appear when the objects in motion aren't properly treated. In order to create good result mosaics, a new method is presented in this paper. The proposed approach uses all the images at the same time to divide the images into different areas by using k-means clustering to create categories, then each category is recommended from the original images with a recommender system. In fact, by considering each image as a user, each pixel as an item, and each item belonging to a category, it is possible to use a recommender system by computing scores with the item profiles. The resulting mosaic will then be a new user to the system. Furthermore, by clustering the images, projection errors are avoided and a better quality mosaic can be created as is seen in the obtained results.

Outdoor images are having several applications including autonomous vehicles, geo-mapping, and surveillance. It is a common phenomenon that the images captured outdoor are prone to noise, which arises due to natural and manmade extreme atmospheric conditions such as haze, fog, and smog. Importantly in autonomous vehicle navigation, it is very important to recover the ground truth image to get the better decision by the system. Estimation of the transmission map and air-light is very crucial in recovering the ground truth image. In this study, the authors proposed a new method to estimate the transmission map based on a mean channel prior (MCP), which represents the depth map to estimate the transmission map. The authors proposed a deep neural network to identify the hazy image for the further dehazing process. In this study, the authors presented, two novel contributions, first an MCP-based image dehazing and second, a deep neural network-based identification of hazy images as a pre-processing block in the proposed end to end system. The proposed deep learning network using the TensorFlow platform provided validation accuracy of 93.4% for hazy image classification. Finally, the proposed MCP-based dehazing network showed better performance in terms of peak-signal-to-noise ratio, structural similarity index, and computational time than that of existing methods.

Rapid advances in artificial intelligence have made it possible to produce forgeries good enough to fool an average user. As a result, there is growing interest in developing robust methods to counter such forgeries. This study presents a new Fourier spectrum-based method for detecting forged text in video images. The authors' premise is that brightness distribution and the spectrum shape exhibit irregular patterns (inconsistencies) for forged text, while appearing more regular for original text. The method divides the spectrum of an input image into sectors and tracks to highlight these effects. Specifically, positive and negative coefficients for sectors and tracks are extracted to quantify the brightness distribution. Variations in the shape of the spectrum are analysed by determining the angular relationship between the principal axes and the sectors/tracks of the spectrum. Next, it combines these two features to detect forged text in the images of IMEI (International Mobile Equipment Identity) numbers and document. For evaluation, the following datasets are used: own video dataset and standard datasets, namely, IMEI number, ICPR 2018 Fraud Document Contest, and a natural scene text dataset. Experimental results show that the proposed method outperforms existing methods in terms of average classification rate and F-score.

Medical image segmentation plays a key role in identifying the disease type. In the last decade, various methods have been proposed for medical images segmentation. Despite many efforts made in medical imaging, segmentation of medical images still faces challenges, concerning the variety of shape, location, and texture quality. According to recent studies and magnetic resonance imaging, segmentation of brain images at around 6 months of age is a challenging issue in brain image segmentation due to low tissue contrast between white matter (WM) and grey matter (GM) regions. In this study, using the deep learning model, the convolutional network for the brain fragmentation is presented. First, the image quality is improved using the pre-processing method. The number of layers utilised in the proposed method is less than that of known models. In the pooling layer, instead of using the maximum function, the averaging function is employed. Sixty-four batches are also considered to improve the performance of the proposed method. The method is evaluated on the iSeg-2017 database. The DISC and ASC measures of the proposed method for the three classes of GM, WM, and cerebrovascular fluid are 0.902, 0.594, 0.930, 0.481, 0.971, and 0.231, respectively.

Currently, the existing literature on multipurpose watermarking employs multiple watermarks for simultaneous achieving of multiple security objectives. This approach can be problematic since a watermark targeting tampers might fail to detect tampering of another watermark inserted for copyright violation. Moreover, most of the current schemes follow a non-blind approach where the original watermark is needed for authentication. A question that naturally arises is whether multipurpose watermarking can be realised with the insertion of a single watermark in a blind way or not. The goal of the authors study is to provide an affirmative answer to this important question. They show how a cryptographic primitive called verifiable threshold secret sharing can be used to come up with a generic construction for blind multipurpose watermarking which inserts a single watermark into the host image for simultaneous achieving of copyright protection, authentication, and tamper localisation. The generic property of the proposed scheme provides flexibility for choosing the embedding/extraction of the watermark based on the desired level of fidelity, robustness and capacity. Experimental results are provided to confirm the superiority of their proposed technique as compared to existing approaches.

Passive millimetre-wave (PMMW) imaging frequently suffers from blurring and low resolution due to the long wavelengths. In addition, the observed images are inevitably disturbed by noise. Traditional image deblurring methods are sensitive to image noise, even a small amount of which will greatly reduce the quality of the point spread function (PSF) estimation. In this paper, we propose a blind deblurring and denoising method via a learning deep denoising convolutional neural networks (DnCNN) denoiser prior and an adaptive -regularized gradient prior for passive millimetre-wave images. First, a blind deblurring restoration model based on the DnCNN denoising prior constraint is established. Second, an adaptive -regularized gradient prior is incorporated into the model to estimate the latent clear image, and the PSF is estimated in the gradient domain. In a multi-scale framework, alternate iterative denoising and deblurring are used to obtain the final PSF estimation and noise estimation. Ultimately, the final clear image is restored by non-blind deconvolution. The experimental results show that the algorithm used in this paper not only has good detail recovery ability but is also more stable to different noise levels. The proposed method is superior to state-of-the-art methods in terms of both subjective measure and visual quality.

The visible spectrum is the most widely used modality for video media. Nonetheless, it is highly dependent on the lighting conditions. Hence, infrared (IR) imaging lower light sensitivity characterisation presents the untapped potential for robust automatic recognition systems. This is applicable to many applications including IR action recognition (AR), which is a relatively young field in IR. As such, in this study, the authors tackle IR and multimodal AR with the proposed utilisation of variational learning of Beta-Liouville (BL) hidden Markov models (HMMs). Furthermore, to the best of the authors' knowledge, this is the first evaluation of the BL HMM in visible AR and in multimodal fusion for AR. They present the results of the proposed model on the infrared action recognition and the IOSB datasets. Experimental results demonstrate promising outcomes. The importance of using IR and multispectral fusion in AR is also highlighted by the results.

Automatic cloud detection algorithm based on supervised learning approach has emerged due to its effectiveness in extracting weather information in satellite images. However, algorithm requires field-expert intervention with huge database of training samples to evaluate its clustering performance. Moreover, lacking in availability of labelled data makes difficult to train the input samples. Therefore, this article puts forward unsupervised many-objective evolutionary clustering technique to discriminate cloudy regions on varying characteristic of underlying surfaces. The study begins with the modification in search capability of -NSGA-III optimisation algorithm by incorporating penalised vector angle concept in associate operator. The analysis of proposed approach has been carried out on benchmark many-objective DTLZ test problems, compared against original -NSGA-III and NSGA-III algorithms. The proposed modified -NSGA-III has been further utilised as clustering technique to solve unsupervised cloud detection problem. Optimal centroid vector for clustering using proposed approach is obtained through modified crossover operator, mutation operator and environmental selection method. Experimental results reveal that proposed approach outperforms comparative many-objective algorithms, MOEA/D and NSGA-III for Landsat 8, MODIS and NOAA satellite images with lower classification average error of % in cloud detection for most of the evaluated test cases.