Automated techniques for Arabic content recognition are at a beginning period contrasted with their partners for the Latin and Chinese contents recognition. There is a bulk of handwritten Arabic archives available in libraries, data centers, historical centers, and workplaces. Digitization of these documents facilitates (1) to preserve and transfer the country’s history electronically, (2) to save the physical storage space, (3) to proper handling of the documents, and (4) to enhance the retrieval of information through the Internet and other mediums. Arabic handwritten character recognition (AHCR) systems face several challenges including the unlimited variations in human handwriting and the leakage of large and public databases. In the current study, the segmentation and recognition phases are addressed. The text segmentation challenges and a set of solutions for each challenge are presented. The convolutional neural network (CNN), deep learning approach, is used in the recognition phase. The usage of CNN leads to significant improvements across different machine learning classification algorithms. It facilitates the automatic feature extraction of images. 14 different native CNN architectures are proposed after a set of try-and-error trials. They are trained and tested on the HMBD database that contains 54,115 of the handwritten Arabic characters. Experiments are performed on the native CNN architectures and the best-reported testing accuracy is 91.96%. A transfer learning (TF) and genetic algorithm (GA) approach named “HMB-AHCR-DLGA” is suggested to optimize the training parameters and hyperparameters in the recognition phase. The pre-trained CNN models (VGG16, VGG19, and MobileNetV2) are used in the later approach. Five optimization experiments are performed and the best combinations are reported. The highest reported testing accuracy is 92.88%.
相似文献This study estimated the effects of multilayer graphene nanoplatelets (MLNGPs) as an additive on polyamide 6 (PA6) spur gear performance. These include strength, elastic modulus, thermal stability, dynamic mechanical analysis, moisture absorption, and wear characteristics.The nanocomposite gear was made by melt mixing method and injection moulded into thick flanges. The flanges were machined using CNC milling machine to produce spur gear. The wear experiments were performed at a running speed of 1400 rpm and at torques of 13 and 16 Nm with different concentration 0, 0.1, 0.3 and 0.5 wt% MLNGPs using test rig. The result showed that 0.3% of MLGNPs is the optimum concentration. Young's modulus increased up to 40%, Vickers microhardness value increased up to 25%, storage modulus E’ is increased up to 37% and glass transition temperature is increased up to 14%. On the other hand TGA result shows that the Tonest increased up to 7.5% and Td increased up to 2%, and wear decreased by 35% at 16 Nm and 54% at 13 Nm. 相似文献
Fibrolamellar hepatocellular carcinoma is a unique malignant liver tumor type which arises in young adults and children. It is uncommon variation subtype of hepatocellular carcinoma which remains ineffectively recorded. Learning of cytogenetic changes in fibrolamellar hepatocellular carcinoma has lagged behind the information obtained from alternate entities of hepatocellular carcinoma lately. Gene expression profiling may prompt new biomarkers that may help develop diagnostic precision for distinguishing fibrolamellar hepatocellular carcinoma. The subatomic cytogenetic approach permits positional identification of gains, amplification, and deletion of DNA sequences of the whole tumor genome, to search for recurrent and particular cytogenetic changes in human fibrolamellar hepatocellular carcinoma. In this work, 13 cell lines of fibrolamellar carcinomas and 30 hepatocellular carcinoma samples examined by a single-nucleotide polymorphs array using two techniques to give more accuracy of the results. The majority of the abnormalities found in the fibrolamellar hepatocellular carcinoma positive cases seen as gain in 1q, 4q, 6q, 7p, 8q, 17q, 20q and loss in 1p, 4p-q, 8p, 11p, 13q, 17p, 18q, 19p, and 22q. The ultimate successive were central amplification at 1q (in 54% of 13 samples), 4q (in 54% of 13 samples), 7p (in 46% of 13 samples), and deletions at 19p13 (in 28% of 13 samples). The study revealed 3 distinct structural variations highlights-related genes MDM4, PRDM5, and WHSC1, and these genes are a novel target signature that can help to predict survival of patients with detecting fibrolamellar hepatocellular carcinoma.
相似文献![点击此处可从《应用聚合物科学杂志》网站下载免费的PDF全文](/ch/ext_images/free.gif)
Reconfigurable sensing antennas (RSA) play a significant role in modern internet-of-things (IOT) applications. The RSAs are capable of transmiting and receiving electromagnetic waves besides sensing different enviroment parameters. This paper introduces a reconfigurable sensing microstrip patch antenna desinged to sense high temperature variations in harsh environment. The indium antimonide (InSb) semiconductor material is a temperature sensitive material employed in RSA designs in the Terahertz (THz) frequency band. An investigation of the temperature dependency of the electrical properties of the InSb-material is introduced. The proposed sensing antenna introduces high sensitivity of 1.588 GHz shift in resonance frequency per unit change in temperature (Kelvin). The resonance frequency of the InSb sensor antenna is changed according to the surrounding environment temperature from 264 GHz to 502.2 GHz with a broadband tuning range of 90.2%. The InSb patch sensor have temperature sensing range of 150 K starting from 250 up-to 400 K. At 300 K the InSb sensor antenna proposes a peak gain of 6.4 dBi with impedance matching bandwidth of 9.57%. An equivalent circuit consists of five lumped elements is estimated for the InSb sensor antenna using particle swarm optimization (PSO) technique at different temperatures. At T?=?250 K the maximum radiation efficiency is 2.5% and is increased up to 86.3% at T?=?400 K.
相似文献