Low-level Feature Extraction Research Articles

Real-time guava tree-part segmentation using fully convolutional network with channel and spatial attention.

It is imminent to develop intelligent harvesting robots to alleviate the burden of rising costs of manual picking. A key problem in robotic harvesting is how to recognize tree parts efficiently without losing accuracy, thus helping the robots plan collision-free paths. This study introduces a real-time tree-part segmentation network by improving fully convolutional network with channel and spatial attention. A lightweight backbone is first deployed to extract low-level and high-level features. These features may contain redundant information in their channel and spatial dimensions, so a channel and spatial attention module is proposed to enhance informative channels and spatial locations. On this basis, a feature aggregation module is investigated to fuse the low-level details and high-level semantics to improve segmentation accuracy. A tree-part dataset with 891 RGB images is collected, and each image is manually annotated in a per-pixel fashion. Experiment results show that when using MobileNetV3-Large as the backbone, the proposed network obtained an intersection-over-union (IoU) value of 63.33 and 66.25% for the branches and fruits, respectively, and required only 2.36 billion floating point operations per second (FLOPs); when using MobileNetV3-Small as the backbone, the network achieved an IoU value of 60.62 and 61.05% for the branches and fruits, respectively, at a speed of 1.18 billion FLOPs. Such results demonstrate that the proposed network can segment the tree-parts efficiently without loss of accuracy, and thus can be applied to the harvesting robots to plan collision-free paths.

A Color-Texture-Based Deep Neural Network Technique to Detect Face Spoofing Attacks

Abstract Given the face spoofing attack, adequate protection of human identity through face has become a significant challenge globally. Face spoofing is an act of presenting a recaptured frame before the verification device to gain illegal access on behalf of a legitimate person with or without their concern. Several methods have been proposed to detect face spoofing attacks over the last decade. However, these methods only consider the luminance information, reflecting poor discrimination of spoofed face from the genuine face. This article proposes a practical approach combining Local Binary Patterns (LBP) and convolutional neural network-based transfer learning models to extract low-level and high-level features. This paper analyzes three color spaces (i.e., RGB, HSV, and YCrCb) to understand the impact of the color distribution on real and spoofed faces for the NUAA benchmark dataset. In-depth analysis of experimental results and comparison with other existing approaches show the superiority and effectiveness of our proposed models.

MM-UNet: A multimodality brain tumor segmentation network in MRI images

The global annual incidence of brain tumors is approximately seven out of 100,000, accounting for 2% of all tumors. The mortality rate ranks first among children under 12 and 10th among adults. Therefore, the localization and segmentation of brain tumor images constitute an active field of medical research. The traditional manual segmentation method is time-consuming, laborious, and subjective. In addition, the information provided by a single-image modality is often limited and cannot meet the needs of clinical application. Therefore, in this study, we developed a multimodality feature fusion network, MM-UNet, for brain tumor segmentation by adopting a multi-encoder and single-decoder structure. In the proposed network, each encoder independently extracts low-level features from the corresponding imaging modality, and the hybrid attention block strengthens the features. After fusion with the high-level semantic of the decoder path through skip connection, the decoder restores the pixel-level segmentation results. We evaluated the performance of the proposed model on the BraTS 2020 dataset. MM-UNet achieved the mean Dice score of 79.2% and mean Hausdorff distance of 8.466, which is a consistent performance improvement over the U-Net, Attention U-Net, and ResUNet baseline models and demonstrates the effectiveness of the proposed model.

A Novel Method for Ground-Based Cloud Image Classification Using Transformer

In recent years, convolutional neural networks (CNNs) have achieved competitive performance in the field of ground-based cloud image (GCI) classification. Proposed CNN-based methods can fully extract the local features of images. However, due to the locality of the convolution operation, they cannot well establish the long-range dependencies between the images, and thus they cannot extract the global features of images. Transformer has been applied to computer vision with great success due to its powerful global modeling capability. Inspired by it, we propose a Transformer-based GCI classification method that combines the advantages of the CNN and Transformer models. Firstly, the CNN model acts as a low-level feature extraction tool to generate local feature sequences of images. Then, the Transformer model is used to learn the global features of the images by efficiently extracting the long-range dependencies between the sequences. Finally, a linear classifier is used for GCI classification. In addition, we introduce a center loss function to address the problem of the simple cross-entropy loss not adequately supervising feature learning. Our method is evaluated on three commonly used datasets: ASGC, CCSN, and GCD. The experimental results show that the method achieves 94.24%, 92.73%, and 93.57% accuracy, respectively, outperforming other state-of-the-art methods. It proves that Transformer has great potential to be applied to GCI classification tasks.

Remote Sensing Image Classification: A Comprehensive Review and Applications

Remote sensing is mainly used to investigate sites of dams, bridges, and pipelines to locate construction materials and provide detailed geographic information. In remote sensing image analysis, the images captured through satellite and drones are used to observe surface of the Earth. The main aim of any image classification-based system is to assign semantic labels to captured images, and consequently, using these labels, images can be arranged in a semantic order. The semantic arrangement of images is used in various domains of digital image processing and computer vision such as remote sensing, image retrieval, object recognition, image annotation, scene analysis, content-based image analysis, and video analysis. The earlier approaches for remote sensing image analysis are based on low-level and mid-level feature extraction and representation. These techniques have shown good performance by using different feature combinations and machine learning approaches. These earlier approaches have used small-scale image dataset. The recent trends for remote sensing image analysis are shifted to the use of deep learning model. Various hybrid approaches of deep learning have shown much better results than the use of a single deep learning model. In this review article, a detailed overview of the past trends is presented, based on low-level and mid-level feature representation using traditional machine learning concepts. A summary of publicly available image benchmarks for remote sensing image analysis is also presented. A detailed summary is presented at the end of each section. An overview regarding the current trends of deep learning models is presented along with a detailed comparison of various hybrid approaches based on recent trends. The performance evaluation metrics are also discussed. This review article provides a detailed knowledge related to the existing trends in remote sensing image classification and possible future research directions.

HCTNet: A Hybrid ConvNet-Transformer Network for Retinal Optical Coherence Tomography Image Classification.

Automatic and accurate optical coherence tomography (OCT) image classification is of great significance to computer-assisted diagnosis of retinal disease. In this study, we propose a hybrid ConvNet-Transformer network (HCTNet) and verify the feasibility of a Transformer-based method for retinal OCT image classification. The HCTNet first utilizes a low-level feature extraction module based on the residual dense block to generate low-level features for facilitating the network training. Then, two parallel branches of the Transformer and the ConvNet are designed to exploit the global and local context of the OCT images. Finally, a feature fusion module based on an adaptive re-weighting mechanism is employed to combine the extracted global and local features for predicting the category of OCT images in the testing datasets. The HCTNet combines the advantage of the convolutional neural network in extracting local features and the advantage of the vision Transformer in establishing long-range dependencies. A verification on two public retinal OCT datasets shows that our HCTNet method achieves an overall accuracy of 91.56% and 86.18%, respectively, outperforming the pure ViT and several ConvNet-based classification methods.

An end-to-end tracking method for polyp detectors in colonoscopy videos

Deep learning based computer-aided diagnosis technology demonstrates an encouraging performance in aspect of polyp lesion detection on reducing the miss rate of polyps during colonoscopies. However, to date, few studies have been conducted for tracking polyps that have been detected in colonoscopy videos, which is an essential and intuitive issue in clinical intelligent video analysis task (e.g. lesion counting, lesion retrieval, report generation). In the paradigm of conventional tracking-by-detection system, detection task for lesion localization is separated from the tracking task for cropped lesions re-identification. In the multi object tracking problem, each target is supposed to be tracked by invoking a tracker after the detector, which introduces multiple inferences and leads to external resource and time consumption. To tackle these problems, we proposed a plug-in module named instance tracking head (ITH) for synchronous polyp detection and tracking, which can be simply inserted into object detection frameworks. It embeds a feature-based polyp tracking procedure into the detector frameworks to achieve multi-task model training. ITH and detection head share the model backbone for low level feature extraction, and then low level feature flows into the separate branches for task-driven model training. For feature maps from the same receptive field, the region of interest head assigns these features to the detection head and the ITH, respectively, and outputs the object category, bounding box coordinates, and instance feature embedding simultaneously for each specific polyp target. We also proposed a method based on similarity metric learning. The method makes full use of the prior boxes in the object detector to provide richer and denser instance training pairs, to improve the performance of the model evaluation on the tracking task. Compared with advanced tracking-by-detection paradigm methods, detectors with proposed ITH can obtain comparative tracking performance but approximate 30% faster speed. Optimized model based on Scaled-YOLOv4 detector with ITH illustrates good trade-off between detection (mAP 91.70%) and tracking (MOTA 92.50% and Rank-1 Acc 88.31%) task at the frame rate of 66 FPS. The proposed structure demonstrates the potential to aid clinicians in real-time detection with online tracking or offline retargeting of polyp instances during colonoscopies.

Joint learning using multiscale attention-enhanced features for remote sensing image scene classification

Scene classification is an important tool for remote sensing image interpretation, and it has fundamental applications in research and industry. However, given complex backgrounds and scale variations, remote sensing images have large intraclass diversity and interclass similarity, which bring challenges to accurate classification of remote sensing images. We proposed a scene classification method using joint learning and multiscale attention to alleviate the aforementioned problems. To fully utilize the multiscale information of the image and improve the adaptability of the proposed method to objects with various sizes, different from general methods that fuse different scales of features for classification, joint learning using multiscale features is developed to optimize the whole network. Specifically, we leverage a pretrained deep convolutional neural network as the feature extractor to extract low-level, medium-level, and high-level feature maps from the images. Then, due to the poor semantics of low-level and medium-level feature maps compared with the high-level feature maps, we design a multiscale attention module to enhance the semantic information and suppress the noise information. Finally, the global mean pooling is used to obtain the feature vectors and different classifiers are used for different feature vectors. And the decision-level fusion is adopted to obtain more reliable predictions. The experimental results on the AID and NWPU-RESISC45 datasets show that the proposed method makes a significant improvement in terms of overall accuracies compared with the baselines. And the overall accuracies of our method on the two datasets are 97.49% and 95.20%, respectively, which achieves state-of-the-art performance. The code will be public at a Github repository available at https://github.com/Cbanyungong/JLMSAF.

Regional Cardiac Motion Scoring With Multi-Scale Motion-Based Spatial Attention.

Regional cardiac motion scoring aims to classify the motion status of each myocardium segment into one of the four categories (normal, hypokinetic, akinetic, and dyskinetic) from multiple short-axis MR sequences. It is essential for prognosis and early diagnosis for various cardiac diseases. However, the complex motion procedure of the myocardium and the invisible pattern differences pose great challenges, leading to low performance for automatic methods. Most existing works mitigate the task by differentiating the normal motion patterns from the abnormal ones, without fine-grained motion scoring. We propose an effective method for the task of cardiac motion scoring by connecting a bottom-up and another top-down branch with a novel motion-based spatial attention module in multi-scale space. Specifically, we use the convolution blocks for low-level feature extraction that acts as a bottom-up mechanism, and the task of optical flow for explicit motion extraction that acts as a top-down mechanism for high-level allocation of spatial attention. To this end, a newly designed Multi-scale Motion-based Spatial Attention (MMSA) module is used as the pivot connecting the bottom-up part and the top-down part, and adaptively weight the low-level features according to the motion information. Experimental results on a newly constructed dataset of 1440 myocardium segments from 90 subjects demonstrate that the proposed MMSA can accurately analyze the regional myocardium motion, with accuracies of 79.3% for 4-way motion scoring, 89.0% for abnormality detection, and correlation of 0.943 for estimation of motion score index. This work has great potential for practical assessmentof cardiac motion function.

I6mA-Caps: a CapsuleNet-based framework for identifying DNA N6-methyladenine sites.

DNA N6-methyladenine (6mA) has been demonstrated to have an essential function in epigenetic modification in eukaryotic species in recent research. 6mA has been linked to various biological processes. It's critical to create a new algorithm that can rapidly and reliably detect 6mA sites in genomes to investigate their biological roles. The identification of 6mA marks in the genome is the first and most important step in understanding the underlying molecular processes, as well as their regulatory functions. In this article, we proposed a novel computational tool called i6mA-Caps which CapsuleNet based a framework for identifying the DNA N6-methyladenine sites. The proposed framework uses a single encoding scheme for numerical representation of the DNA sequence. The numerical data is then used by the set of convolution layers to extract low-level features. These features are then used by the capsule network to extract intermediate-level and later high-level features to classify the 6mA sites. The proposed network is evaluated on three datasets belonging to three genomes which are Rosaceae, Rice and Arabidopsis thaliana. Proposed method has attained an accuracy of 96.71%, 94% and 86.83% for independent Rosaceae dataset, Rice dataset and A.thaliana dataset respectively. The proposed framework has exhibited improved results when compared with the existing top-of-the-line methods. A user-friendly web-server is made available for the biological experts which can be accessed at: http://nsclbio.jbnu.ac.kr/tools/i6mA-Caps/. Supplementary data are available at Bioinformatics online.

Improving vegetation segmentation with shadow effects based on double input networks using polarization images

Fractional vegetation cover (FVC) plays an important role in the study of vegetation growth state, and the key issue is accurately segmenting and extracting green vegetation from the background. However, the shadows generated by natural lights produce extreme illuminance differences in images, which greatly reduces the vegetation extraction accuracy. The polarization information for ground objects is independent of the physical state of the reflectivity of ground objects, and it can be used to eliminate the influence of strong reflections in images to a certain extent, reduce the illuminance differences under extreme sunlight conditions, and help improve the vegetation recognition effect under shadow conditions. To improve the accuracy of vegetation segmentation under shadow conditions, this study introduces polarized reflection information for vegetation and an improved semantic segmentation network, notably a double input residual network based on DeepLabv3plus (DIR_DeepLabv3plus), with fusion strategies based on concatenation and addition is proposed. The network extracts low-level features and high-level features at different spatial scales from both light intensity (red-greenblue (RGB)) images and degree of linear polarization (DoLP) images independently by a deep residual network and atrous spatial pyramid pooling (ASPP) structure, effectively improving the accuracy of vegetation segmentation in shadow situations. The results show that the mean intersection over union (mIoU) values of vegetation without shadows, with light shadows and with shadows are 94.01%, 92.508% and 90.969%, respectively. Compared with the color index method and green fractional vegetation cover extraction from digital images using a shadow-resistant algorithm (SHAR-LABFVC), the proposed method provides a greatly improved extraction accuracy, and it has 0.18%, 1.00% and 1.49% higher mIoU values for vegetation under different shadow conditions than the method without polarization information. This study provides a new approach for vegetation segmentation and improves the accuracy of FVC calculations under shadow conditions.

A Classification Model with Cognitive Reasoning Ability

In this paper, we study the classification problem of large data with many features and strong feature dependencies. This type of problem has shortcomings when handled by machine learning models. Therefore, a classification model with cognitive reasoning ability is proposed. The core idea is to use cognitive reasoning mechanism proposed in this paper to solve the classification problem of large structured data with multiple features and strong correlation between features, and then implements cognitive reasoning for features. The model has three parts. The first part proposes a Feature-to-Image algorithm for converting structured data into image data. The algorithm quantifies the dependencies between features, so as to take into account the impact of individual independent features and correlations between features on the prediction results. The second part designs and implements low-level feature extraction of the quantified features using convolutional neural networks. With the relative symmetry of the capsule network, the third part proposes a cognitive reasoning mechanism to implement high-level feature extraction, feature cognitive reasoning, and classification tasks of the data. At the same time, this paper provides the derivation process and algorithm description of cognitive reasoning mechanism. Experiments show that our model is efficient and outperforms comparable models on the category prediction experiment of ADMET properties of five compounds.This work will provide a new way for cognitive computing of intelligent data analysis.

Memory-inspired spiking hyperdimensional network for robust online learning

Recently, brain-inspired computing models have shown great potential to outperform today’s deep learning solutions in terms of robustness and energy efficiency. Particularly, Spiking Neural Networks (SNNs) and HyperDimensional Computing (HDC) have shown promising results in enabling efficient and robust cognitive learning. Despite the success, these two brain-inspired models have different strengths. While SNN mimics the physical properties of the human brain, HDC models the brain on a more abstract and functional level. Their design philosophies demonstrate complementary patterns that motivate their combination. With the help of the classical psychological model on memory, we propose SpikeHD, the first framework that fundamentally combines Spiking neural network and hyperdimensional computing. SpikeHD generates a scalable and strong cognitive learning system that better mimics brain functionality. SpikeHD exploits spiking neural networks to extract low-level features by preserving the spatial and temporal correlation of raw event-based spike data. Then, it utilizes HDC to operate over SNN output by mapping the signal into high-dimensional space, learning the abstract information, and classifying the data. Our extensive evaluation on a set of benchmark classification problems shows that SpikeHD provides the following benefit compared to SNN architecture: (1) significantly enhance learning capability by exploiting two-stage information processing, (2) enables substantial robustness to noise and failure, and (3) reduces the network size and required parameters to learn complex information.

A Context Feature Enhancement Network for Building Extraction from High-Resolution Remote Sensing Imagery

The complexity and diversity of buildings make it challenging to extract low-level and high-level features with strong feature representation by using deep neural networks in building extraction tasks. Meanwhile, deep neural network-based methods have many network parameters, which take up a lot of memory and time in training and testing. We propose a novel fully convolutional neural network called the Context Feature Enhancement Network (CFENet) to address these issues. CFENet comprises three modules: the spatial fusion module, the focus enhancement module, and the feature decoder module. First, the spatial fusion module aggregates the spatial information of low-level features to obtain buildings’ outline and edge information. Secondly, the focus enhancement module fully aggregates the semantic information of high-level features to filter the information of building-related attribute categories. Finally, the feature decoder module decodes the output of the above two modules to segment the buildings more accurately. In a series of experiments on the WHU Building Dataset and the Massachusetts Building Dataset, our CFENet balances efficiency and accuracy compared to the other four methods we compared, and achieves optimality on all five evaluation metrics: PA, PC, F1, IoU, and FWIoU. This indicates that CFENet can effectively enhance and fuse buildings’ low-level and high-level features, improving building extraction accuracy.

Multi-channel Calibrated Transformer with Shifted Windows for few-shot fault diagnosis under sharp speed variation

In engineering practice, mechanical equipment is mainly operated under the working conditions of sharp speed variations, which results the data distribution domain shift. Furthermore, the domain shift and the lack of data in engineering practice render severe challenges for existing intelligent mechanical faults diagnosis technologies. To this end, this paper proposed a Multi-channel Calibrated Transformer with Shifted Windows (MCSwin-T) for computing self-attention in each non-overlapping window which models the relations between the sequences of split patches. Meanwhile, a new partitioning approach is designed by shifting the windows and alternately use the two different partitioning approach to establish the connections across windows. To extract low-level features of the signal and maintain the positional information, a plurality of convolution layers is applied before transformer block. A normalized method which is a multi-channel multiplication of the vector generated by each residual block is also developed to calibrate activation and increase the stability of the optimization. To evaluate the effectiveness, the proposed method is compared with multiple advanced transformer methods in two case studies under speed transient conditions. The experimental results indicate the superiority and higher accuracy of the proposed method under few-shot domain shift condition.

Land Use Classification Method of Remote Sensing Images for Urban and Rural Planning Monitoring Using Deep Learning

Aiming at the problems that most existing segmentation methods are difficult to deal with the imbalance of remote sensing image distribution and the overlap of segmentation target edges, a land use classification method of remote sensing image for urban and rural planning monitoring based on deep learning is proposed. Firstly, the U-Net is improved by pooling index upsampling and dimension superposition. The improvement can not only extract high-level abstract features but also extract low-level detail features, so as to reduce the loss of image edge information in the process of deconvolution. Then, the batch normalization and scaling exponential linear unit (SeLU) are used to improve the U-Net model. Finally, the improved U-Net model is applied to the classification of remote sensing images of land use types to realize dynamic monitoring. The experimental analysis of the proposed method based on TensorFlow deep learning framework shows that its total accuracy exceeds 94%. The segmentation effect of land use types in remote sensing images is good.

Hybrid Architecture Based on CNN and Transformer for Strip Steel Surface Defect Classification

Strip steel surface defects occur frequently during the manufacturing process, and these defects cause hidden risks in the use of subsequent strip products. Therefore, it is crucial to classify the strip steel’s surface defects accurately and efficiently. Most classification models of strip steel surface defects are generally based on convolutional neural networks (CNNs). However, CNNs, with local receptive fields, do not have admirable global representation ability, resulting in poor classification performance. To this end, we proposed a hybrid network architecture (CNN-T), which merges CNN and Transformer encoder. The CNN-T network has both strong inductive biases (e.g., translation invariance, locality) and global modeling capability. Specifically, CNN first extracts low-level and local features from the images. The Transformer encoder then globally models these features, extracting abstract and high-level semantic information and finally sending them to the multilayer perceptron classifier for classification. Extensive experiments show that the classification performance of CNN-T outperforms pure Transformer networks and CNNs (e.g., GoogLeNet, MobileNet v2, ResNet18) on the NEU-CLS dataset (training ratio is 80%) with a 0.28–2.23% improvement in classification accuracy, with fewer parameters (0.45 M) and floating-point operations (0.12 G).

Vision-based apple quality grading with multi-view spatial network

• Describing overall features by capturing apple images from multiple perspectives. • Lightweight CNN helps model more practical. • Aggregating spatial apple features is realized by bilateral long short-term memory. • The performance and efficiency are far excellent with a 99.23% accuracy. Apple grading is one of the typical tasks in food grading. For the image-based apple grading task, the variety of apple shapes and the incompleteness of single-view visual information pose great difficulties in achieving high accuracy of apple grading. This paper proposes a novel multi-view spatial network to tackle the apple grading task, which incorporates apple size information as one of the grading criteria, to solve those problems that affect the accuracy of grading. Specifically, we first use well-pretrained lightweight CNNs to extract low-level features of each view of apple. Second, we formulate a spatial feature aggregation module through a bidirectional LSTM and mean-pooling to exploit the correlative information of apple from multiple views. Besides, we elaborately design a multi-view apple data collection scheme for apple grading dataset construction. The experimental results indicate that the accuracy of Multi-View Spatial Network reaches 99.23%, which is quite promising in apple grading task and outperforms some state-of-the-art grading methods.

Effective of Modern Techniques on Content-Based Medical Image Retrieval: A Survey

The advancement in medical imaging has resulted in a rapid and large increase in medical images inside repositories. These medical images contain very important information that can be used in many things, including diagnosing diseases. This implies that a precise, efficient way of indexing and retrieving biomedical images is necessary to obtain medical images from such repositories in real-time. CBMIR, therefore, played an important part, where the CBMIR’s area is very important in the field of image processing and involves low-level feature extraction and similarity measures for the comparison of medical images such as color histograms, edges, texture, shape. The majority of the methods already in use in CBMIR enhance the retrieval of a medical image and diseases diagnosis by reducing the issue of the semantic gap between low visual and high semantic levels. Also, secure access to the medical image of diverse cases, which are often kept on a network and are susceptible to malicious attacks is considered an important target for all medical practitioners. So, most CBMIRs try to cover this target for the purpose of privacy preservation. So, in this survey, the most advanced (CBMIR) frameworks that were used to reduce the issue of semantic gaps, high dimensionality feature maps were covered, disease diagnosis, and medical image security. Furthermore, the different publicly and standard databases used in measuring the performance of these frameworks also were covered.

Deep Learning Approach with Optimization Algorithm for Reducing the Training and Testing Time in SAR Image Detection and Recognition

Objective: - To reduce training and testing time in SAR image detection and recognition using optimization algorithm. Methods: SAR images have achieved a prominent position in the arena of remote sensing and satellite technology. SAR image processing has many applications in different areas like agriculture, mineral exploration, resource management and environmental monitoring. To carry out this research works, MSTAR dataset was were used with six classes. The collected dataset was preprocessed with filtering algorithms to remove the speckle noise from the image. Then, image segmentation which is essential expertise for image processing has been done with aim of rationalizing and changing image representation into more meaningful and easier to analyze. The characters of Hue, Intensity, Saturation (H, I, S) were applied to acquire the information of the pixels of the target image. By doing so, color information and edge extraction were done, since it was the basic idea to achieve the segmented image from its background. Next, feature extraction has been done using DNNs through three stages (Low, middle and high level feature extraction). At low level feature extraction, the image edge and lines were extracted while image front or noses were extracted at middle level feature extraction. Then, all image features were combined at the high level feature extraction and thus all features were combined to form high-level features, because they were primitive images features. After doing all the above, detection has been done to locate the presence of objects in an image using bounding box regression model. Finally, for the SAR image recognition, the pre-trained CNN models such as ResNet-50, AlexNet, and VGG16 were used to compare their performance with the proposed model. In SAR image detection and recognition, the high training and testing time is founded as a challenging. Thus, to reduce such long training and testing time of SAR image detection and recognition, optimization algorithms such as Stochastic Gradient Descent with Momentum (SGDM), RMSProp and Adam optimization methods were used with the pre-training and proposed CNN model. Findings: Preprocessing was carried out using Median, Guided Filter (GF), Lee, Box, Adaptive or Wiener filter algorithms were used and their performances were also compared in PSNR, SNR and MSE values and from those all used algorithms, the GF achieves better performance in high PSNR value of 37.8342. The performance of the all three pre-trained models and the proposed models were compared in accuracy and speed. The AlexNet, ResNet-50, VGG16 and proposed models achieved accuracy of 89%, 92%, 86% and 95% respectively and proposed model achieved the best performance. Among the used four models with Optimization methods, the Proposed Model with SGDM took very least time for Training (26’ and 49s) and for Testing (17s). Novelty: New Deep Network Model was successfully designed, developed and used along with Optimization algorithms for reducing the training and testing time in SAR image detection and recognition. Keywords: ResNet50; VGG16; SAR image; Optimization algorithms