Narrow Results

With the development of information technology and the increasing demand for data processing, the serial mode of the central processing unit (CPU) is difficult to efficiently transmit large-scale spatiotemporal data, and the processing effect for high-resolution images is not good. This paper designed a high-resolution image processing and spatiotemporal data transmission system based on graphics processing unit (GPU) acceleration to improve the processing efficiency of large-scale spatiotemporal data. In this paper, traffic spatiotemporal data was taken as an example for analysis. Large-scale traffic image data was collected by road monitoring equipment, and image compression was performed on the collected image. Fourier transform was used to eliminate image data redundancy, and GPU-accelerated parallel processing was used to achieve fast image defogging and data transmission. This paper selected 2TB of traffic spatiotemporal data with image resolutions of 540P, 720P, 1080P, 1440P, and 2160P. GPU acceleration was performed using the Compute Unified Device Architecture (CUDA). In images with a resolution of 2160P, the processing time for CPU and GPU acceleration was 2900ms and 28ms, respectively, with an acceleration ratio of 103.6. A high-resolution image processing and spatiotemporal data transmission system based on GPU acceleration can improve the efficiency of traffic spatiotemporal data processing and have excellent concurrent processing capabilities.

In the context of artificial intelligence, machine vision technology has also become a research hotspot in the decoration industry. However, there are still many problems in the practical application process of engineering. Mainly affected by uncertain factors such as environmental noise, there is still no comprehensive method to determine the treatment plan for decoration pattern structure accurately. Neural network technology has been applied to various scenarios for recognition and has achieved good results. This paper provides a reference for balancing the accuracy and speed of the model by controlling the model parameters for sampling the characteristic structure of decoration patterns. Pre-train the pruned model using an embedded system and fully utilize a software-defined lightweight training model as a benchmark. Compared with traditional neural networks, it has the characteristics of a flexible structure, high computational efficiency, and strong adaptability. Moreover, based on the benchmark model for quantification, the experimental results of MobileNet and Adam quantization were compared. The feature recognition rate and computational cost were 8.4% and 11.3% lower than the comparison scheme, respectively, improving the efficiency and quality of image recognition. Optimization algorithms are more precise and have unique machine learning analysis capabilities, which can help enhance pattern recognition in the decoration industry and provide references for similar recognition needs in other industries.

Breast cancer (BrC) is one of the most common causes of death among women worldwide. Images of the breast (mammography or ultrasound) may show an anomaly that represents early indicators of BrC. However, accurate breast image interpretation necessitates labor-intensive procedures and highly skilled medical professionals. As a second opinion for the physician, deep learning (DL) tools can be useful for the diagnosis and classification of malignant and benign lesions. However, due to the lack of interpretability of DL algorithms, it is not easy to understand by experts as to how to predict a label. In this work, we proposed multitask U-Net Saliency estimation and DL model-based breast lesion segmentation and classification using ultrasound images. A new contrast enhancement technique is proposed to improve the quality of original images. After that, a new technique was proposed called UNET-Saliency map for the segmentation of breast lesions. Simultaneously, a MobileNetV2 deep model is fine-tuned with additional residual blocks and trained from scratch using original and enhanced images. The purpose of additional blocks is to reduce the number of parameters and better learning of ultrasound images. Training is performed from scratch and extracted features from the deeper layers of both models. In the later step, a new cross-entropy controlled sine-cosine algorithm is developed and selected best features. The main purpose of this step is the reduction of irrelevant features for the classification phase. The selected features are fused in the next step by employing a serial-based Manhattan Distance (SbMD) approach and classified the resultant vector using machine learning classifiers. The results indicate that a wide neural network (W-NN) obtained the highest accuracy of 98.9% and sensitivity rate of 98.70% on the selected breast ultrasound image dataset. The comparison of the proposed method accuracy is conducted with state-of-the-art (SoArt) techniques which show the improved performance.

Food security is a national strategic plan for Saudi Arabia and a part of Saudi Vision 2030. Focusing on an increased area under production to reduce import dependency on foods and, ultimately, food security has led to higher domestic production of food crops. Pepper is cultivated worldwide, and several farmer’s subsistence depends on these crops. Unfortunately, due to lower pepper productivity, farmers involved in pepper cultivation face enormous losses caused by various pepper diseases. These losses can be avoided if the disease is detected timely and accurate. The time needed for the procedure and improper detection cannot help reduce the losses and cannot be released from the diseases. Deep learning (DL)-based automated techniques are one of the most computing techniques can be used in environmental modeling for detecting plant disease can provide promising outcomes to the users for acquiring higher accuracy within a shorter time for recognizing pepper diseases. This study designs an Automated Pepper Leaf Disease Recognition and Classification using Optimal Deep Learning (APLDRC-ODL) technique to improve sustainable agriculture in KSA. The purpose of the APLDRC-ODL technique is to enhance crop productivity and reduce crop losses in KSA via the detection of pepper leaf diseases. The APLDRC-ODL technique involves a multi-faceted approach comprising median filtering (MF)-based noise elimination and Otsu thresholding-based segmentation as a preprocessing step. At the same time, complex and intrinsic features can be generated by the utilization of the capsule network (CapsNet) model. Meanwhile, the African vulture fractals optimization algorithm (AVOA) can be applied for the optimal hyperparameter selection process. Lastly, an extreme learning machine (ELM) classifier is utilized for the detection and classification of pepper diseases. An extensive set of experiments is performed to highlight the efficiency of the APLDRC-ODL technique under the pepper leaf disease dataset. The simulation process of the APLDRC-ODL technique highlighted a superior value of 98.32% compared to existing models.

Agriculture is India’s most common job, yet it lacks innovation and technology. As the world’s population expands, so does the demand for more food. Pesticides are used on farms to boost yield. The toxicity of the fertilizer has serious health repercussions for the farmer. So, it’s recommended to measure the amount of pesticide used and only apply it when necessary. We devised an insect-finding and insecticide-spraying mechanism. This is accomplished by employing a drone or Uninterrupted Ariel Vehicle. The drone has a camera that can photograph fields and lift pesticides weighing 3 to 4kg. After locating the insect, the insecticide is sprayed through the nozzles. In the proposed model, the Deep Convolutional Neural Network (CNN) has reached state of the art in image processing and object detection issues. Deep CNN has the potential to self-learn hidden features that help with insect detection. When compared to other similar approaches, experimental findings on a real dataset to illustrate the usefulness of the suggested methodology. We identified insects on the crop with 90% accuracy using deep CNN. It helps farmers to increase crop yield while also shielding them from the detrimental effects of spraying pesticides on the field manually.

Skin cancer is described as an abnormal, exponential growth of skin cells that originate from melanocytes due to DNA impair or damage. It is threatening because of its ability to metastasize to other body parts. Early diagnosis of melanoma can lessen the morbidity and mortality associated to skin cancer. Subjective visual inspection of melanoma may vary among investigators due to scarcity of medical tools and different level of experience. Therefore, accurate lesion detection becomes a tedious and time-consuming job. Several computer-aided diagnosis (CAD) systems based on machine and deep learning models have emerged in-order to assist clinicians in timely diagnosis of malignant (cancerous) melanoma. The conventional approaches derive low-level, handcrafted features from dermoscopy images of skin lesion. Novel deep learning-based neural networks are developed which aim to extract more generic and deep features for model training. Moreover, dermoscopy and clinical images play a central role in accurate detection of cancerous melanoma. This review paper is organized in five steps: First, we explain image analysis techniques of lesion images. Second, we highlight the challenges in identifying a lesion as melanoma or benign. Third, we provide an overview of publicly available skin lesion datasets. Fourth, we review the performance of various machine and deep learning-based melanoma diagnosis frameworks. This study exhibits that deep learning-based models and their ensemble outperform conventional machine learning approaches in respect of reliability, accuracy and sensitivity. The future work may be define bout the advisement of deep leaning algorithm.

This paper considers a memory system which provides simultaneous accesses to several image points within a linesegment in a 2-dimensional image array. Image processing tasks such as line detection with Hough transform, rotation, or line accesses in an image array can benefit from the proposed memory system with a single processor such that the overall memory access time is reduced.

Image processing applications are suitable candidates for parallelism and have at least in part motivated the design and development of some of the pioneering massively parallel processing systems including the CLIP family, the DAP, the MPP and the GAPP. In this paper, we describe the implementation of various image processing algorithms on the MasPar massively parallel computer system. The suitability of the MasPar for solving image processing algorithms is demonstrated either by parallelizing the algorithms using successful known techniques and/or developing new techniques suitable for the MasPar architecture. We quantitatively evaluate the performance of MasPar in solving these problems. Then, we compare its performance to various related massively parallel architectures. It is shown that the MasPar system compares favorably to these architectures, and is able to execute many fundamental image processing applications in a time amenable to real-time processing. Thus, the MasPar seems to be a promising architecture for massively parallel real-time image processing applications.

The computational model on which the algorithms are developed is the array with reconfigurable optical buses (AROB). It integrates the advantages of both optical transmission and electronic computation. The main contributions of this paper are in designing several optimal and/or optimal speed-up template matching algorithms with varying degrees of parallelism on the AROB model. For an N × N digitized image and an M × M template, when the domains of the image and the template are O(log N)-bit integers, we first design several basic operations for window broadcasting and rotation. Then based on these basic operations, three efficient and scalable algorithms for template matching are derived using various numbers of processors on a two-dimensional (2-D) or 3-D AROB. For 1 ≤ r ≤ N, 1 ≤ p ≤ M ≤ q ≤ N, one runs in time using r × r processors, another runs in , (resp. ) time using pN × pN/log M (resp. pN × pN × log N) processors, and the other runs in (resp. ) time using pq × pq/log M (or pq × pqN × log N) processors, respectively. The latter two algorithms can be tuned to run in O(1) time on a 2-D AROB. To the best of our knowledge, there are no algorithms which can reach this time complexity for this problem on a 2-D array architecture.

The main contribution of this paper is to show a new approach for FM screening which we call Local Exhaustive Search (LES) method, and to present ways to accelerate the computation using an FPGA. FM screening, as opposed to conventional AM screening, keeps unit dot size when converting an original gray-scale image into the binary image for printing. FM screening pays great attention to generate moiré-free binary images reproducing continuous-tone and fine details of original photographic images. Our basic approach for FM screening is to generate a binary image whose projected image onto human eyes is very close to the original image. The projected image is computed by applying a Gaussian filter to the binary image. LES performs an exhaustive search for each of the small square subimages in the binary image and replaces the subimage by the best binary pattern. The exhaustive search is repeated until no more improvement is possible. The experimental results show that LES produces a high quality and sharp binary image. We also implemented LES on an FPGA to accelerate the computation and achieved a speedup factor of up to 51 over the software implementations.

Screening is an important task to convert a continuous-tone image into a binary image with pure black and white pixels. The main contribution of this paper is to show a new algorithm for cluster-dot screening using a local exhaustive search. Our new algorithm generates 2-cluster, 3-cluster, and 4-cluster binary images, in which all dots have at least 2, 3, and 4 pixels, respectively. The key idea of our new screening method is to repeat a local exhaustive search that finds the best binary pattern in small windows of size k × k in a binary image. The experimental results show that the local exhaustive search produces high quality and sharp cluster-dot binary images. We also present an hardware algorithm to accelerate the computation. Our hardware algorithm for a round of the local exhaustive search runs O(k²) clock cycles while the software implementation runs in O(2^k2 w²) time, where (2w + 1) × (2w + 1) is the size of Gaussian filter. Thus, from theoretical point of view, our hardware algorithm achieves a speedup factor of O(w²). To show that our hardware algorithm is practically fast, we have implemented it on an FPGA. Our hardware algorithm achieved a speedup factor of up to 229 over the software implementation.

Connected component labeling is a process that assigns unique labels to the connected components of a binary image. The main contribution of this paper is to present a low-latency hardware connected component labeling algorithm for k-concave binary images designed and implemented in FPGA. Pixels of a binary image are given to the FPGA in raster order, and the resulting labels are also output in the same order. The advantage of our labeling algorithm is low latency and to use a small internal storage of the FPGA. We have implemented our hardware labeling algorithm in an Altera Stratix Family FPGA, and evaluated the performance. The implementation result shows that for a 10-concave binary image of 2048 × 2048, our connected component labeling algorithm runs in approximately 70ms and its latency is approximately 750µs.

Single image super-resolution (SISR) aims to obtain a high-resolution output from one low-resolution image. Currently, deep learning-based SISR approaches have been widely discussed in medical image processing, because of their potential to achieve high-quality, high spatial resolution images without the cost of additional scans. However, most existing methods are designed for scale-specific SR tasks and are unable to generalize over magnification scales. In this paper, we propose an approach for medical image arbitrary-scale super-resolution (MIASSR), in which we couple meta-learning with generative adversarial networks (GANs) to super-resolve medical images at any scale of magnification in $(1,4]$. Compared to state-of-the-art SISR algorithms on single-modal magnetic resonance (MR) brain images (OASIS-brains) and multi-modal MR brain images (BraTS), MIASSR achieves comparable fidelity performance and the best perceptual quality with the smallest model size. We also employ transfer learning to enable MIASSR to tackle SR tasks of new medical modalities, such as cardiac MR images (ACDC) and chest computed tomography images (COVID-CT). The source code of our work is also public. Thus, MIASSR has the potential to become a new foundational pre-/post-processing step in clinical image analysis tasks such as reconstruction, image quality enhancement, and segmentation.

The prevalence of dementia is currently increasing worldwide. This syndrome produces a deterioration in cognitive function that cannot be reverted. However, an early diagnosis can be crucial for slowing its progress. The Clock Drawing Test (CDT) is a widely used paper-and-pencil test for cognitive assessment in which an individual has to manually draw a clock on a paper. There are a lot of scoring systems for this test and most of them depend on the subjective assessment of the expert. This study proposes a computer-aided diagnosis (CAD) system based on artificial intelligence (AI) methods to analyze the CDT and obtain an automatic diagnosis of cognitive impairment (CI). This system employs a preprocessing pipeline in which the clock is detected, centered and binarized to decrease the computational burden. Then, the resulting image is fed into a Convolutional Neural Network (CNN) to identify the informative patterns within the CDT drawings that are relevant for the assessment of the patient’s cognitive status. Performance is evaluated in a real context where patients with CI and controls have been classified by clinical experts in a balanced sample size of $3282$ drawings. The proposed method provides an accuracy of $75.65\%$ in the binary case-control classification task, with an AUC of $0.83$. These results are indeed relevant considering the use of the classic version of the CDT. The large size of the sample suggests that the method proposed has a high reliability to be used in clinical contexts and demonstrates the suitability of CAD systems in the CDT assessment process. Explainable artificial intelligence (XAI) methods are applied to identify the most relevant regions during classification. Finding these patterns is extremely helpful to understand the brain damage caused by CI. A validation method using resubstitution with upper bound correction in a machine learning approach is also discussed.

Parkinson’s Disease (PD) is the second most prevalent neurodegenerative disorder among adults. Although its triggers are still not clear, they may be due to a combination of different types of biomarkers measured through medical imaging, metabolomics, proteomics or genetics, among others. In this context, we have proposed a Computer-Aided Diagnosis (CAD) system that combines structural and functional imaging data from subjects in Parkinson’s Progression Markers Initiative dataset by means of an Ensemble Learning methodology trained to identify and penalize input sources with low classification rates and/ or high-variability. This proposal improves results published in recent years and provides an accurate solution not only from the point of view of image preprocessing (including a comparison between different intensity preservation techniques), but also in terms of dimensionality reduction methods (Isomap). In addition, we have also introduced a bagging classification schema for scenarios with unbalanced data.

As shown by our results, the CAD proposal is able to detect PD with $96.48\%$ of balanced accuracy, and opens up the possibility of combining any number of input data sources relevant for PD.

Spiking neural membrane systems (SN P systems) are a class of bio-inspired models inspired by the activities and connectivity of neurons. Extensive studies have been made on SN P systems with synchronization-based communication, while further efforts are needed for the systems with rhythm-based communication. In this work, we design an asynchronous SN P system with resonant connections where all the enabled neurons in the same group connected by resonant connections should instantly produce spikes with the same rhythm. In the designed system, each of the three modules implements one type of the three operations associated with the edge detection of digital images, and they collaborate each other through the resonant connections. An algorithm called EDSNP for edge detection is proposed to simulate the working of the designed asynchronous SN P system. A quantitative analysis of EDSNP and the related methods for edge detection had been conducted to evaluate the performance of EDSNP. The performance of the EDSNP in processing the testing images is superior to the compared methods, based on the quantitative metrics of accuracy, error rate, mean square error, peak signal-to-noise ratio and true positive rate. The results indicate the potential of the temporal firing and the proper neuronal connections in the SN P system to achieve good performance in edge detection.

In this work, engineered nanostructures (ENS) have been fabricated on the packed integrated circuits. Coding lookup tables were developed to assign different digits in numerical matrices to different fabricated nano-signatures. The numerical matrices are encrypted according to advanced encryption standard (AES). The encrypted numerical matrix is ink printed on the components, and the nanosignatures are fabricated on the packaged of the chips via electron beam lithography (EBL). This process is to be done in the manufacturer side of the supply chain. The numerical matrix and the nanosignature accompany the product in its long journey in the global supply chain. The global supply chain is proved to be susceptible to counterfeiters. For keeping counterfeiters‘ hands out of the process, the cipher key and the coding lookup tables are provided to the consumer using a secure direct line between the authentic manufacturer and the consumer. In the consumer side, the printed numerical matrix is decrypted. Having the decrypted numerical matrix makes it possible to extract the nanosignature from the laser speckle pattern shined on the packaged product. In this work, an algorithm is developed to extract the nano-signature by having the decrypted matrix and reflected laser speckle patterns as inputs. Confirming the existence of the nano-signature confirms the authenticity of the component. Imitating the nano-signatures by the counterfeiters is not possible because there is no way for them to observe the shape of these signatures without having access to the cipher key.

In this paper we address a relaxation theorem for a new integral functional of the calculus of variations defined on the space of functions in whose gradient is an L^p-vector field with distributional divergence given by a Radon measure. The result holds for integrand of type f(x, Δu) without any coerciveness condition, with respect to the second variable, and C¹-continuity assumptions with respect to the spatial variable x.

In this paper, we investigate the performance of a Computer Aided Diagnosis (CAD) system for the detection of clustered microcalcifications in mammograms. Our detection algorithm consists of the combination of two different methods. The first, based on difference-image techniques and gaussianity statistical tests, finds out the most obvious signals. The second, is able to discover more subtle microcalcifications by exploiting a multiresolution analysis by means of the wavelet transform. We can separately tune the two methods, so that each one of them is able to detect signals with similar features. By combining signals coming out from the two parts through a logical OR operation, we can discover microcalcifications with different characteristics. Our algorithm yields a sensitivity of 91.4% with 0.4 false positive cluster per image on the 40 images of the Nijmegen database.

We have developed a method for the detection of clusters of microcalcifications in digital mammograms. Here, we present a genetic algorithm used to optimize the choice of the parameters in the detection scheme. The optimization has allowed the improvement of the performance, the detailed study of the influence of the various parameters on the performance and an accurate investigation of the behavior of the detection method on unknown cases. We reach a sensitivity of 96.2% with 0.7 false positive clusters per image on the Nijmegen database; we are also able to identify the most significant parameters. In addition, we have examined the feasibility of a distributed genetic algorithm implemented on a non-dedicated Cluster Of Workstations. We get very good results both in terms of quality and efficiency.