Inception Block Research Articles

VGGIN-Net: Deep Transfer Network for Imbalanced Breast Cancer Dataset.

In this paper, we have presented a novel deep neural network architecture involving transfer learning approach, formed by freezing and concatenating all the layers till block4 pool layer of VGG16 pre-trained model (at the lower level) with the layers of a randomly initialized naïve Inception block module (at the higher level). Further, we have added the batch normalization, flatten, dropout and dense layers in the proposed architecture. Our transfer network, called VGGIN-Net, facilitates the transfer of domain knowledge from the larger ImageNet object dataset to the smaller imbalanced breast cancer dataset. To improve the performance of the proposed model, regularization was used in the form of dropout and data augmentation. A detailed block-wise fine tuning has been conducted on the proposed deep transfer network for images of different magnification factors. The results of extensive experiments indicate a significant improvement of classification performance after the application of fine-tuning. The proposed deep learning architecture with transfer learning and fine-tuning yields the highest accuracies in comparison to other state-of-the-art approaches for the classification of BreakHis breast cancer dataset. The articulated architecture is designed in a way that it can be effectively transfer learned on other breast cancer datasets.

Interpatient Heartbeat Classification Using Modified Residual Attention Network With Two-Phase Training and Assistant Decision

State-of-the-art studies on automatic heartbeat classification have made efforts to improve supraventricular ectopic beat (SVEB) detection, but resulted in a low positive predictive value (PPV) of SVEB and poor performance of other categories under the interpatient paradigm. This article proposes a novel algorithm for heartbeat classification by the means of deep learning combined with feature extraction, which enhances the performance and can be applied to an automatic electrocardiogram (ECG) analysis. It is an improved residual attention network (IraNet) that combines attention module, residual inception block, and bidirectional long short-term memory (BiLSTM) layer to process single heartbeat segments directly. A two-phase training method is presented to change the weights in a planned way to deal with the imbalance of distinguishable features. An assistant decision for the deep learning-based model enhances the PPV of SVEB effectively. Under the interpatient paradigm on the Massachusetts Institute of Technology and Beth Israel Hosipital (MIT-BIH) arrhythmia database, the overall accuracy (Acc) of four classes achieves 0.9548. For SVEB class, the PPV is significantly improved to 0.7557 with the sensitivity (Sen) and an <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$F1$ </tex-math></inline-formula> score of 0.7392 and 0.7474, respectively. For ventricular ectopic beat (VEB) class, the Sen, PPV, and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$F1$ </tex-math></inline-formula> score achieve 0.9575, 0.8500, and 0.9005, respectively, and the numbers are 0.9663, 0.9873, and 0.9767 for normal beat class. For fusion beat class, the numbers are 0.7964, 0.3597, and 0.4954, respectively. The proposed algorithm gets competitive results with the state-of-the-art studies, and there is a notable improvement in the detection of SVEB

Soft measurement of oil–water two-phase flow using a multi-task sequence-based CapsNet

Flow parameters measurement facilitates the understanding of two-phase flow. Due to the changeable structures of the flow, the prediction of superficial velocity of oil–water two-phase flow in large diameter pipes is still a challenging problem. Therefore, we first conducted a vertical upward oil–water two-phase flow experiment in a 125 mm ID pipe, and obtained the response signals under different flow conditions by a vertical multi-electrode array (VMEA) conductance sensor. Then, new data pre-processing (1D to 2D) techniques and information fusion techniques (network channels) are employed. Moreover, the front-end structure of the network is optimized using a combination of attention block and residual structure, and the middle structure is optimized using inception block; on the other hand, the back-end structure of the original capsule network is innovatively changed so that it can handle both the flow pattern classification and superficial velocity prediction tasks. The dynamic routing algorithm has also been improved to speed up model training. Extensive experiments validate the effectiveness of the improved modules. Finally, we compare the proposed network with its variants and other competing networks. The better performance results show that our multi-task sequence-based CapsNet has great potential for dealing with high-dimensional, time-varying and nonlinear problems in multiphase flow.

Generative Adversarial Network Combined with SE-ResNet and Dilated Inception Block for Segmenting Retinal Vessels.

This study develops an accurate method based on the generative adversarial network (GAN) that targets the issue of the current discontinuity of micro vessel segmentation in the retinal segmentation images. The processing of images has become increasingly efficient since the advent of deep learning method. We have proposed an improved GAN combined with SE-ResNet and dilated inception block for the segmenting retinal vessels (SAD-GAN). The GAN model has been improved with respect to the following points. (1) In the generator, the original convolution block is replaced with SE-ResNet module. Furthermore, SE-Net can extract the global channel information, while concomitantly strengthening and weakening the key features and invalid features, respectively. The residual structure can alleviate the issue of gradient disappearance. (2) The inception block and dilated convolution are introduced into the discriminator, which enhance the transmission of features and expand the acceptance domain for improved extraction of the deep network features. (3) We have included the attention mechanism in the discriminator for combining the local features with the corresponding global dependencies, and for highlighting the interdependent channel mapping. SAD-GAN performs satisfactorily on public retina datasets. On DRIVE dataset, ROC_AUC and PR_AUC reach 0.9813 and 0.8928, respectively. On CHASE_DB1 dataset, ROC_AUC and PR_AUC reach 0.9839 and 0.9002, respectively. Experimental results demonstrate that the generative adversarial model, combined with deep convolutional neural network, enhances the segmentation accuracy of the retinal vessels far above that of certain state-of-the-art methods.

Deep Compressive Sensing on ECG Signals with Modified Inception Block and LSTM.

In practical electrocardiogram (ECG) monitoring, there are some challenges in reducing the data burden and energy costs. Therefore, compressed sensing (CS) which can conduct under-sampling and reconstruction at the same time is adopted in the ECG monitoring application. Recently, deep learning used in CS methods improves the reconstruction performance significantly and can removes of some of the constraints in traditional CS. In this paper, we propose a deep compressive-sensing scheme for ECG signals, based on modified-Inception block and long short-term memory (LSTM). The framework is comprised of four modules: preprocessing; compression; initial; and final reconstruction. We adaptively compressed the normalized ECG signals, sequentially using three convolutional layers, and reconstructed the signals with a modified Inception block and LSTM. We conducted our experiments on the MIT-BIH Arrhythmia Database and Non-Invasive Fetal ECG Arrhythmia Database to validate the robustness of our model, adopting Signal-to-Noise Ratio (SNR) and percentage Root-mean-square Difference (PRD) as the evaluation metrics. The PRD of our scheme was the lowest and the SNR was the highest at all of the sensing rates in our experiments on both of the databases, and when the sensing rate was higher than 0.5, the PRD was lower than 2%, showing significant improvement in reconstruction performance compared to the comparative methods. Our method also showed good recovering quality in the noisy data.

A Novel Deep Learning Model for Sea State Classification Using Visual-Range Sea Images

Wind-waves exhibit variations both in shape and steepness, and their asymmetrical nature is a well-known feature. One of the important characteristics of the sea surface is the front-back asymmetry of wind-wave crests. The wind-wave conditions on the surface of the sea constitute a sea state, which is listed as an essential climate variable by the Global Climate Observing System and is considered a critical factor for structural safety and optimal operations of offshore oil and gas platforms. Methods such as statistical representations of sensor-based wave parameters observations and numerical modeling are used to classify sea states. However, for offshore structures such as oil and gas platforms, these methods induce high capital expenditures (CAPEX) and operating expenses (OPEX), along with extensive computational power and time requirements. To address this issue, in this paper, we propose a novel, low-cost deep learning-based sea state classification model using visual-range sea images. Firstly, a novel visual-range sea state image dataset was designed and developed for this purpose. The dataset consists of 100,800 images covering four sea states. The dataset was then benchmarked on state-of-the-art deep learning image classification models. The highest classification accuracy of 81.8% was yielded by NASNet-Mobile. Secondly, a novel sea state classification model was proposed. The model took design inspiration from GoogLeNet, which was identified as the optimal reference model for sea state classification. Systematic changes in GoogLeNet’s inception block were proposed, which resulted in an 8.5% overall classification accuracy improvement in comparison with NASNet-Mobile and a 7% improvement from the reference model (i.e., GoogLeNet). Additionally, the proposed model took 26% less training time, and its per-image classification time remains competitive.

Automatic Cobb angle measurement method based on vertebra segmentation by deep learning.

The accuracy of the Cobb measurement is essential for the diagnosis and treatment of scoliosis. Manual measurement is however influenced by the observer variability hence affecting progression evaluation. In this paper, we propose a fully automatic Cobb measurement method to address the accuracy issue of manual measurement. We improve the U-shaped network based on the multi-scale feature fusion to segment each vertebra. To enable multi-scale feature extraction, the convolution kernel of the U-shaped network is substituted by the Inception Block. To solve the problem of gradient disappearance caused by the widening of the network structure from the Inception Block, we propose using ResBlock. CBAM (Convolutional Block Attention Module) can help the network judges the importance of the feature map to modify learning weight. Also, to further enhance the accuracy of feature extraction, we add the CBAM to the U-shaped network bottleneck. Finally, based on the segmented vertebrae, the efficient automatic Cobb angle measurement method is proposed to estimate the Cobb angle. In the experiments, 75 spinal X-ray images are tested. We compare the proposed U-Shaped network with the state-of-the-art methods including DeepLabV3 + , FCN8S, SegNet, U-Net, U-Net + + , BASNet, and U2Net for vertebra segmentation. Our results show that compared to these methods, the Dice coefficient is improved by 32.03%, 33.58%, 12.42%, 5.65%, 4.55%, 4.42%, and 3.27%, respectively. The CMAE of the calculated Cobb measurement is 2.45°, which is lower than the average error of 5-7° of manual measurement. The experimental results indicate that the improved U-shaped network improves the accuracy of vertebra segmentation. The proposed efficient automatic Cobb measurement method can be used in clinics to reduce observer variability.

Application of deep learning in structural health management of concrete structures

Structural health management constitutes an essential factor in ensuring the durability of concrete structures. Cracks in concrete, reinforcement corrosion, alkali–silica reaction and efflorescence attacks are common concrete defects that can be identified visually. However, the detection and classification of these defects in concrete bridges and other high-rise concrete structures is a difficult and expensive process using manual approaches. In this research, a deep learning application is applied to detect and classify concrete defects. Concrete images from the public repository were used to create the explored database. The database was divided into training and validation subsets. The visual geometry group (Vgg19), neural search architecture (nasnetlarge) and residual inception block (vinceptionresnetv2) algorithms were used to analyse the images. The results of the overall performance show that the Vgg19 algorithm recorded higher accuracy in the detection and classification of concrete defects when compared to the nasnetlarge and inceptionresnetv2 algorithms. The efficiency of the proposed approach was evaluated using a new dataset containing images of concrete defects. The outcome of this research shows clearly that deep learning models could enhance the efficiency of concrete structural health monitoring in a multi-classification scenario.

Analyzing Effect on Residual Learning by Gradual Narrowing Fully-Connected Layer Width and Implementing Inception Block in Convolution Layer

Analyzing Effect on Residual Learning by Gradual Narrowing Fully-Connected Layer Width and Implementing Inception Block in Convolution Layer

Fault Detection of Aero-Engine Sensor Based on Inception-CNN

The aero-engine system is complex, and the working environment is harsh. As the fundamental component of the aero-engine control system, the sensor must monitor its health status. Traditional sensor fault detection algorithms often have many parameters, complex architecture, and low detection accuracy. Aiming at this problem, a convolutional neural network (CNN) whose basic unit is an inception block composed of convolution kernels of different sizes in parallel is proposed. The network fully extracts redundant analytical information between sensors through different size convolution kernels and uses it for aero-engine sensor fault detection. On the sensor failure dataset generated by the Monte Carlo simulation method, the detection accuracy of Inception-CNN is 95.41%, which improves the prediction accuracy by 17.27% and 12.69% compared with the best-performing non-neural network algorithm and simple BP neural networks tested in the paper, respectively. In addition, the method simplifies the traditional fault detection unit composed of multiple fusion algorithms into one detection algorithm, which reduces the complexity of the algorithm. Finally, the effectiveness and feasibility of the method are verified in two aspects of the typical sensor fault detection effect and fault detection and isolation process.

Gradient-Guided and Multi-Scale Feature Network for Image Super-Resolution

Recently, deep-learning-based image super-resolution methods have made remarkable progress. However, most of these methods do not fully exploit the structural feature of the input image, as well as the intermediate features from the intermediate layers, which hinders the ability of detail recovery. To deal with this issue, we propose a gradient-guided and multi-scale feature network for image super-resolution (GFSR). Specifically, a dual-branch structure network is proposed, including the trunk branch and the gradient one, where the latter is used to extract the gradient feature map as structural prior to guide the image reconstruction process. Then, to absorb features from different layers, two effective multi-scale feature extraction modules, namely residual of residual inception block (RRIB) and residual of residual receptive field block (RRRFB), are proposed and embedded in different network layers. In our RRIB and RRRFB structures, an adaptive weighted residual feature fusion block (RFFB) is investigated to fuse the intermediate features to generate more beneficial representations, and an adaptive channel attention block (ACAB) is introduced to effectively explore the dependencies between channel features to further boost the feature representation capacity. Experimental results on several benchmark datasets demonstrate that our method achieves superior performance against state-of-the-art methods in terms of both subjective visual quality and objective quantitative metrics.

Intelligent Security Performance Prediction for IoT-Enabled Healthcare Networks Using an Improved CNN

The global healthcare industry and artificial intelligence have promoted the development of the diversified intelligent healthcare applications. Internet of Things (IoT) will play an important role in meeting the high throughput requirements of diversified intelligent healthcare applications. However, the mobile IoT-enabled healthcare networks are diverse and open, the healthcare big data transmission is vulnerable to a potential attack, which can cause network outages and serious healthcare security issues. To process the complex healthcare security event in real time, security performance prediction is critical for mobile IoT-enabled healthcare networks. In this article, we first analyze the security performance, and derive the novel expressions for the security performance in a closed form. Then, to analyze the security performance in real time, a security performance intelligent prediction algorithm is proposed. An improved convolutional neural network (CNN) model is designed, which combines the four-layer convolution and a four-branch inception block, and can adopt different convolution kernels in the same layer. The four-branch inception block can increase the width of the CNN while reducing the parameters. The improved CNN model can not only increases the width of the CNN, extract different sizes of healthcare data features, but also increases the adaptability to the nonlinear healthcare big data. Compared with different methods, the proposed intelligent algorithm can obtain better security performance prediction. In particular, for prediction precision, the proposed intelligent algorithm is increased by 20%.

A Deep Shared Multi-Scale Inception Network Enables Accurate Neonatal Quiet Sleep Detection With Limited EEG Channels.

In this paper, we introduce a new variation of the Convolutional Neural Network Inception block, called Sinc, for sleep stage classification in premature newborn babies using electroencephalogram (EEG). In practice, there are many medical centres where only a limited number of EEG channels are recorded. Existing automated algorithms mainly use multi-channel EEGs which perform poorly when fewer numbers of channels are available. The proposed Sinc utilizes multi-scale analysis to place emphasis on the temporal EEG information to be less dependent on the number of EEG channels. In Sinc, we increase the receptive fields through Inception while by additionally sharing the filters that have similar receptive fields, overfitting is controlled and the number of trainable parameters dramatically reduced. To train and test this model, 96 longitudinal EEG recordings from 26 premature infants are used. The Sinc-based model significantly outperforms state-of-the-art neonatal quiet sleep detection algorithms, with mean Kappa 0.77 ± 0.01 (with 8-channel EEG) and 0.75 ± 0.01 (with a single bipolar channel EEG). This is the first study using Inception-based networks for EEG analysis that utilizes filter sharing to improve efficiency and trainability. The suggested network can successfully detect quiet sleep stages with even a single EEG channel making it more practical especially in the hospital setting where cerebral function monitoring is predominantly used.

Mixer U-Net: An Improved Automatic Road Extraction from UAV Imagery

Automatic road extraction from unmanned aerial vehicle (UAV) imagery has been one of the major research topics in the area of remote sensing analysis due to its importance in a wide range of applications such as urban planning, road monitoring, intelligent transportation systems, and automatic road navigation. Thanks to the recent advances in Deep Learning (DL), the tedious manual segmentation of roads can be automated. However, the majority of these models are computationally heavy and, thus, are not suitable for UAV remote-sensing tasks with limited resources. To alleviate this bottleneck, we propose two lightweight models based on depthwise separable convolutions and ConvMixer inception block. Both models take the advantage of computational efficiency of depthwise separable convolutions and multi-scale processing of inception module and combine them in an encoder–decoder architecture of U-Net. Specifically, we substitute standard convolution layers used in U-Net for ConvMixer layers. Furthermore, in order to learn images on different scales, we apply ConvMixer layer into Inception module. Finally, we incorporate pathway networks along the skip connections to minimize the semantic gap between encoder and decoder. In order to validate the performance and effectiveness of the models, we adopt Massachusetts roads dataset. One incarnation of our models is able to beat the U-Net’s performance with 10× fewer parameters, and DeepLabV3’s performance with 12× fewer parameters in terms of mean intersection over union (mIoU) metric. For further validation, we have compared our models against four baselines in total and used additional metrics such as precision (P), recall (R), and F1 score.

Adaptive spectral-spatial feature fusion network for hyperspectral image classification using limited training samples

Recently, the excellent power of spectral-spatial feature representation of convolutional neural network (CNN) has gained widespread attention for hyperspectral image (HSI) classification. Nevertheless, the practical performance of CNN-based models in HSI classification is ordinarily limited by the available amount of the training samples. In this article, we investigate the limitations of current CNN-based methods for HSI feature (spectral and spatial) extraction and utilization. For spectral features, the distant inter-band relationships are often neglected. Therefore, a novel spectral band non-localization (SBNL) operation is proposed to enable the non-local spectral inter-band correlations to be excavated by convolutional kernels with limited receptive fields. For spatial features, the extracted spatial multiscale features conventionally isolated in different channels. Subsequently, we develop a novel multiscale-share inception block (MSIB) to exploit the cross-relationships among the multiscale features. More significantly, to better take advantage of the complementary information of spectral and spatial features, a plug-and-play adaptive feature fusion (AFF) module is introduced. Eventually, the adaptive spectral-spatial feature fusion network (AS2F2N) is introduced for HSI classification. Experimental results derived from three benchmark data sets exhibit that the proposed method outperforms previous state-of-the-art CNN-based methods under limited training samples situation. The codes of this work will be available at https://github.com/zhonghaocheng/ELSEVIER_IJAEOG_AS2F2N for the sake of reproducibility.

Potato Plant Leaf Diseases Identification Using Transfer Learning

Background/Objectives: Agriculture is a major food source for Ethiopian population. Plant diseases contribute a great production loss, which can be addressed with continuous monitoring. Early plant disease identification using computer vision and Artificial Intelligence (AI) helps the farmers to take preventive course of action to increase production quality. Manual plant disease identification is strenuous and error-prone. Methods: In this study, we present a convolutional neural network architecture inception-v3 model to detect potato leaf diseases using a deep learning-based transfer learning technique. We used separable convolution in the inception block that can minimize the number of parameters by an outsized margin and to utilize resource efficiently. The inception-V3 model have a higher training accuracy and needs less training time than the main CNN architecture, as the used parameters are fewer. Findings: In this study, there is an improvement on the little noisy on sample images which leads to misidentification of diseases. In our experiment, we have used an RGB color channel image dataset to train model, which yields an overall accuracy performance of 98.7% on the heldout test set. Novelty: In order to identify potato leave diseases, we conducted transfer learning for high performance classification with pixel-wise operation to enhance the number of leaf images. A model based on inception-v3 transfer learning approach is presented in this study for disease identification of potato leave images, thus provide an effective computer-aided recognition model for potato disease classification in the absence of large data. Keywords: Artificial intelligence; convolutional neural network; deep learning; leaf disease identification; Softmax

Multimodal Imaging of Target Detection Algorithm under Artificial Intelligence in the Diagnosis of Early Breast Cancer.

This study aimed to analyze the diagnostic value of multimodal images based on artificial intelligence target detection algorithms for early breast cancer, so as to provide help for clinical imaging examinations of breast cancer. This article combined residual block with inception block, constructed a new target detection algorithm to detect breast lumps, used deep convolutional neural network and ultrasound imaging in diagnosing benign and malignant breast lumps, took breast density grading with mammography, compared the convolutional neural network (CNN) algorithm with the proposed algorithm, and then applied the proposed algorithm to the diagnosis of 120 female patients with breast lumps. According to the results, accuracy rates of breast lump detection (94.76%), benign and malignant breast lumps diagnosis (98.22%), and breast grading (93.65%) with the algorithm applied in this study were significantly higher than those (75.67%, 87.23%, and 79.54%) with CNN algorithm, and the difference was statistically significant (P < 0.05); among 62 patients with malignant breast lumps of the 120 patients with breast lumps, 37 were patients with invasive ductal carcinoma, 8 with lobular carcinoma in situ, 16 with intraductal carcinoma, and 4 with mucinous carcinoma; among the remaining 58 patients with benign breast lumps, 28 were patients with fibrocystic breast disease, 17 with intraductal papilloma, 4 with breast hyperplasia, and 9 with adenopathy; the differences in shape, growth direction, edge, and internal echo of multimodal ultrasound imaging of patients with benign and malignant breast lumps had statistical significance (P < 0.05); the malignant constituent ratios of patients with breast density grades I to IV were 0%, 7.10%, 80.40%, and 100%, respectively. In short, the multimodal imaging diagnosis under the algorithm in this article was superior to CNN algorithm in all aspects; according to the judgment on benign and malignant breast lumps and breast density with multimodal imaging features, the higher the breast density, the higher the probability of breast cancer.

Accurate Identification Strategy of Coal and Gangue Using Infrared Imaging Technology Combined With Convolutional Neural Network

To effectively separate coal and gangue, accurate classification is an important prerequisite. Here, a new recognition solution for coal and gangue is proposed, in which the convolutional neural network (CNN) is trained to achieve the automatically identifying coal and gangue based on the infrared images without considering the selection of feature extraction and classifier. Firstly, the specific architecture and detailed parameters of the model are optimized and the CNN model based on only one Inception Block contains three different convolution kernels are considered to be the most appropriate model. Next, performance of the proposed identification model is analyzed and evaluated by the infrared image dataset, and we discovered that the CNN model is capable of correctly identifying 192 training samples and 48 test samples. Finally, compared with the traditional recognition model and other CNN recognition model, it is proved that the proposed CNN model has superior recognition performance. The results state clearly that the combination of infrared image and CNN can quickly and accurately identify coal and gangue without complex image processing steps. At the same time, the model has a certain anti-interference ability for different noises. And it has a certain reference value for the research and development of intelligent coal preparation equipment.

Ultrasound spine image segmentation using multi-scale feature fusion Skip-Inception U-Net (SIU-Net)

Scoliosis is a 3D spinal deformation where the spine takes a lateral curvature, forming an angle in the coronal plane. Diagnosis of scoliosis requires periodic detection, and frequent exposure to radiative imaging may cause cancer. A safer and more economical alternative imaging, i.e., 3D ultrasound imaging modality, is being explored. However, unlike other radiative modalities, an ultrasound image is noisy, which often suppresses the image's useful information. Through this research, a novel hybridized CNN architecture, multi-scale feature fusion Skip-Inception U-Net (SIU-Net), is proposed for a fully automatic bony feature detection, which can be further used to assess the severity of scoliosis safely and automatically. The proposed architecture, SIU-Net, incorporates two novel features into the basic U-Net architecture: (a) an improvised Inception block and (b) newly designed decoder-side dense skip pathways. The proposed model is tested on 109 spine ultrasound image datasets. The architecture is evaluated using the popular (i) Jaccard Index (ii) Dice Coefficient and (iii) Euclidean distance, and compared with (a) the basic U-net segmentation model, (b) a more evolved UNet++ model, and (c) a newly developed MultiResUNet model. The results show that SIU-Net gives the clearest segmentation output, especially in the important regions of interest such as thoracic and lumbar bony features. The method also gives the highest average Jaccard score of 0.781 and Dice score of 0.883 and the lowest histogram Euclidean distance of 0.011 than the other three models. SIU-Net looks promising to meet the objectives of a fully automatic scoliosis detection system.

Light-Weight Localization and Scale-Independent Multi-gate UNET Segmentation of Left and Right Ventricles in MRI Images.

Heart segmentation in cardiac magnetic resonance images is heavily used during the assessment of left ventricle global function. Automation of the segmentation is crucial to standardize the analysis. This study aims at developing a CNN-based framework to aid the clinical measurements of the left ventricle and right ventricle in cardiac magnetic resonance images. We propose a fully automated framework for localization and segmentation of the left ventricle and right ventricle in both short- and long-axis views from cardiac magnetic resonance images. The localization module utilizes a light-weight model that detects the region of interest and feeds it to the segmentation model. Also, we propose the Multi-Gate block as an extension to the UNet to boost the segmentation performance by aggregating multi-scale features. Comparison between our proposed method and the baseline UNet was performed to show the gain in the overall performance. The reliability of the model was assessed by testing the method against cardiac magnetic resonance images with different levels of noise and deformations. Heart localization accuracy was 0.59 and 1.75 pixels in both short- and long-axis views respectively. Left and right ventricle blood-pool segmentation Dice was (0.93, 0.90) in end-systole and (0.97, 0.95) in end-diastole. The left ventricle myocardium was segmented accurately with Dice of 0.91 and 0.90 in end-systole and end-diastole respectively. Left ventricle ejection fraction was found to be highly correlated with the gold standard with r = 0.987. Moreover, the proposed pipeline is fast, achieving 0.002 sec per image on average. Adding the Multi-Gate Dilated Inception Block has boosted the performance of UNet architecture and has shown generalization ability when tested on noisy and deformed cardiac magnetic resonance images. The proposed method has proven its wide applicability and reliability for heart detection when tested on different datasets.