Search results for: AlexNet
Commenced in January 2007
Frequency: Monthly
Edition: International
Paper Count: 10

Search results for: AlexNet

10 The Modification of Convolutional Neural Network in Fin Whale Identification

Authors: Jiahao Cui

Abstract:

In the past centuries, due to climate change and intense whaling, the global whale population has dramatically declined. Among the various whale species, the fin whale experienced the most drastic drop in number due to its popularity in whaling. Under this background, identifying fin whale calls could be immensely beneficial to the preservation of the species. This paper uses feature extraction to process the input audio signal, then a network based on AlexNet and three networks based on the ResNet model was constructed to classify fin whale calls. A mixture of the DOSITS database and the Watkins database was used during training. The results demonstrate that a modified ResNet network has the best performance considering precision and network complexity.

Keywords: convolutional neural network, ResNet, AlexNet, fin whale preservation, feature extraction

Procedia PDF Downloads 126
9 Loss Function Optimization for CNN-Based Fingerprint Anti-Spoofing

Authors: Yehjune Heo

Abstract:

As biometric systems become widely deployed, the security of identification systems can be easily attacked by various spoof materials. This paper contributes to finding a reliable and practical anti-spoofing method using Convolutional Neural Networks (CNNs) based on the types of loss functions and optimizers. The types of CNNs used in this paper include AlexNet, VGGNet, and ResNet. By using various loss functions including Cross-Entropy, Center Loss, Cosine Proximity, and Hinge Loss, and various loss optimizers which include Adam, SGD, RMSProp, Adadelta, Adagrad, and Nadam, we obtained significant performance changes. We realize that choosing the correct loss function for each model is crucial since different loss functions lead to different errors on the same evaluation. By using a subset of the Livdet 2017 database, we validate our approach to compare the generalization power. It is important to note that we use a subset of LiveDet and the database is the same across all training and testing for each model. This way, we can compare the performance, in terms of generalization, for the unseen data across all different models. The best CNN (AlexNet) with the appropriate loss function and optimizers result in more than 3% of performance gain over the other CNN models with the default loss function and optimizer. In addition to the highest generalization performance, this paper also contains the models with high accuracy associated with parameters and mean average error rates to find the model that consumes the least memory and computation time for training and testing. Although AlexNet has less complexity over other CNN models, it is proven to be very efficient. For practical anti-spoofing systems, the deployed version should use a small amount of memory and should run very fast with high anti-spoofing performance. For our deployed version on smartphones, additional processing steps, such as quantization and pruning algorithms, have been applied in our final model.

Keywords: anti-spoofing, CNN, fingerprint recognition, loss function, optimizer

Procedia PDF Downloads 137
8 An Evaluation of Neural Network Efficacies for Image Recognition on Edge-AI Computer Vision Platform

Authors: Jie Zhao, Meng Su

Abstract:

Image recognition, as one of the most critical technologies in computer vision, works to help machine-like robotics understand a scene, that is, if deployed appropriately, will trigger the revolution in remote sensing and industry automation. With the developments of AI technologies, there are many prevailing and sophisticated neural networks as technologies developed for image recognition. However, computer vision platforms as hardware, supporting neural networks for image recognition, as crucial as the neural network technologies, need to be more congruently addressed as the research subjects. In contrast, different computer vision platforms are deterministic to leverage the performance of different neural networks for recognition. In this paper, three different computer vision platforms – Jetson Nano(with 4GB), a standalone laptop(with RTX 3000s, using CUDA), and Google Colab (web-based, using GPU) are explored and four prominent neural network architectures (including AlexNet, VGG(16/19), GoogleNet, and ResNet(18/34/50)), are investigated. In the context of pairwise usage between different computer vision platforms and distinctive neural networks, with the merits of recognition accuracy and time efficiency, the performances are evaluated. In the case study using public imageNets, our findings provide a nuanced perspective on optimizing image recognition tasks across Edge-AI platforms, offering guidance on selecting appropriate neural network structures to maximize performance under hardware constraints.

Keywords: alexNet, VGG, googleNet, resNet, Jetson nano, CUDA, COCO-NET, cifar10, imageNet large scale visual recognition challenge (ILSVRC), google colab

Procedia PDF Downloads 92
7 6D Posture Estimation of Road Vehicles from Color Images

Authors: Yoshimoto Kurihara, Tad Gonsalves

Abstract:

Currently, in the field of object posture estimation, there is research on estimating the position and angle of an object by storing a 3D model of the object to be estimated in advance in a computer and matching it with the model. However, in this research, we have succeeded in creating a module that is much simpler, smaller in scale, and faster in operation. Our 6D pose estimation model consists of two different networks – a classification network and a regression network. From a single RGB image, the trained model estimates the class of the object in the image, the coordinates of the object, and its rotation angle in 3D space. In addition, we compared the estimation accuracy of each camera position, i.e., the angle from which the object was captured. The highest accuracy was recorded when the camera position was 75°, the accuracy of the classification was about 87.3%, and that of regression was about 98.9%.

Keywords: 6D posture estimation, image recognition, deep learning, AlexNet

Procedia PDF Downloads 157
6 Identification of Breast Anomalies Based on Deep Convolutional Neural Networks and K-Nearest Neighbors

Authors: Ayyaz Hussain, Tariq Sadad

Abstract:

Breast cancer (BC) is one of the widespread ailments among females globally. The early prognosis of BC can decrease the mortality rate. Exact findings of benign tumors can avoid unnecessary biopsies and further treatments of patients under investigation. However, due to variations in images, it is a tough job to isolate cancerous cases from normal and benign ones. The machine learning technique is widely employed in the classification of BC pattern and prognosis. In this research, a deep convolution neural network (DCNN) called AlexNet architecture is employed to get more discriminative features from breast tissues. To achieve higher accuracy, K-nearest neighbor (KNN) classifiers are employed as a substitute for the softmax layer in deep learning. The proposed model is tested on a widely used breast image database called MIAS dataset for experimental purposes and achieved 99% accuracy.

Keywords: breast cancer, DCNN, KNN, mammography

Procedia PDF Downloads 137
5 Malignancy Assessment of Brain Tumors Using Convolutional Neural Network

Authors: Chung-Ming Lo, Kevin Li-Chun Hsieh

Abstract:

The central nervous system in the World Health Organization defines grade 2, 3, 4 gliomas according to the aggressiveness. For brain tumors, using image examination would have a lower risk than biopsy. Besides, it is a challenge to extract relevant tissues from biopsy operation. Observing the whole tumor structure and composition can provide a more objective assessment. This study further proposed a computer-aided diagnosis (CAD) system based on a convolutional neural network to quantitatively evaluate a tumor's malignancy from brain magnetic resonance imaging. A total of 30 grade 2, 43 grade 3, and 57 grade 4 gliomas were collected in the experiment. Transferred parameters from AlexNet were fine-tuned to classify the target brain tumors and achieved an accuracy of 98% and an area under the receiver operating characteristics curve (Az) of 0.99. Without pre-trained features, only 61% of accuracy was obtained. The proposed convolutional neural network can accurately and efficiently classify grade 2, 3, and 4 gliomas. The promising accuracy can provide diagnostic suggestions to radiologists in the clinic.

Keywords: convolutional neural network, computer-aided diagnosis, glioblastoma, magnetic resonance imaging

Procedia PDF Downloads 147
4 Comparison of Deep Convolutional Neural Networks Models for Plant Disease Identification

Authors: Megha Gupta, Nupur Prakash

Abstract:

Identification of plant diseases has been performed using machine learning and deep learning models on the datasets containing images of healthy and diseased plant leaves. The current study carries out an evaluation of some of the deep learning models based on convolutional neural network (CNN) architectures for identification of plant diseases. For this purpose, the publicly available New Plant Diseases Dataset, an augmented version of PlantVillage dataset, available on Kaggle platform, containing 87,900 images has been used. The dataset contained images of 26 diseases of 14 different plants and images of 12 healthy plants. The CNN models selected for the study presented in this paper are AlexNet, ZFNet, VGGNet (four models), GoogLeNet, and ResNet (three models). The selected models are trained using PyTorch, an open-source machine learning library, on Google Colaboratory. A comparative study has been carried out to analyze the high degree of accuracy achieved using these models. The highest test accuracy and F1-score of 99.59% and 0.996, respectively, were achieved by using GoogLeNet with Mini-batch momentum based gradient descent learning algorithm.

Keywords: comparative analysis, convolutional neural networks, deep learning, plant disease identification

Procedia PDF Downloads 200
3 Land Cover Remote Sensing Classification Advanced Neural Networks Supervised Learning

Authors: Eiman Kattan

Abstract:

This study aims to evaluate the impact of classifying labelled remote sensing images conventional neural network (CNN) architecture, i.e., AlexNet on different land cover scenarios based on two remotely sensed datasets from different point of views such as the computational time and performance. Thus, a set of experiments were conducted to specify the effectiveness of the selected convolutional neural network using two implementing approaches, named fully trained and fine-tuned. For validation purposes, two remote sensing datasets, AID, and RSSCN7 which are publicly available and have different land covers features were used in the experiments. These datasets have a wide diversity of input data, number of classes, amount of labelled data, and texture patterns. A specifically designed interactive deep learning GPU training platform for image classification (Nvidia Digit) was employed in the experiments. It has shown efficiency in training, validation, and testing. As a result, the fully trained approach has achieved a trivial result for both of the two data sets, AID and RSSCN7 by 73.346% and 71.857% within 24 min, 1 sec and 8 min, 3 sec respectively. However, dramatic improvement of the classification performance using the fine-tuning approach has been recorded by 92.5% and 91% respectively within 24min, 44 secs and 8 min 41 sec respectively. The represented conclusion opens the opportunities for a better classification performance in various applications such as agriculture and crops remote sensing.

Keywords: conventional neural network, remote sensing, land cover, land use

Procedia PDF Downloads 372
2 Systematic Evaluation of Convolutional Neural Network on Land Cover Classification from Remotely Sensed Images

Authors: Eiman Kattan, Hong Wei

Abstract:

In using Convolutional Neural Network (CNN) for classification, there is a set of hyperparameters available for the configuration purpose. This study aims to evaluate the impact of a range of parameters in CNN architecture i.e. AlexNet on land cover classification based on four remotely sensed datasets. The evaluation tests the influence of a set of hyperparameters on the classification performance. The parameters concerned are epoch values, batch size, and convolutional filter size against input image size. Thus, a set of experiments were conducted to specify the effectiveness of the selected parameters using two implementing approaches, named pertained and fine-tuned. We first explore the number of epochs under several selected batch size values (32, 64, 128 and 200). The impact of kernel size of convolutional filters (1, 3, 5, 7, 10, 15, 20, 25 and 30) was evaluated against the image size under testing (64, 96, 128, 180 and 224), which gave us insight of the relationship between the size of convolutional filters and image size. To generalise the validation, four remote sensing datasets, AID, RSD, UCMerced and RSCCN, which have different land covers and are publicly available, were used in the experiments. These datasets have a wide diversity of input data, such as number of classes, amount of labelled data, and texture patterns. A specifically designed interactive deep learning GPU training platform for image classification (Nvidia Digit) was employed in the experiments. It has shown efficiency in both training and testing. The results have shown that increasing the number of epochs leads to a higher accuracy rate, as expected. However, the convergence state is highly related to datasets. For the batch size evaluation, it has shown that a larger batch size slightly decreases the classification accuracy compared to a small batch size. For example, selecting the value 32 as the batch size on the RSCCN dataset achieves the accuracy rate of 90.34 % at the 11th epoch while decreasing the epoch value to one makes the accuracy rate drop to 74%. On the other extreme, setting an increased value of batch size to 200 decreases the accuracy rate at the 11th epoch is 86.5%, and 63% when using one epoch only. On the other hand, selecting the kernel size is loosely related to data set. From a practical point of view, the filter size 20 produces 70.4286%. The last performed image size experiment shows a dependency in the accuracy improvement. However, an expensive performance gain had been noticed. The represented conclusion opens the opportunities toward a better classification performance in various applications such as planetary remote sensing.

Keywords: CNNs, hyperparamters, remote sensing, land cover, land use

Procedia PDF Downloads 170
1 Small Scale Mobile Robot Auto-Parking Using Deep Learning, Image Processing, and Kinematics-Based Target Prediction

Authors: Mingxin Li, Liya Ni

Abstract:

Autonomous parking is a valuable feature applicable to many robotics applications such as tour guide robots, UV sanitizing robots, food delivery robots, and warehouse robots. With auto-parking, the robot will be able to park at the charging zone and charge itself without human intervention. As compared to self-driving vehicles, auto-parking is more challenging for a small-scale mobile robot only equipped with a front camera due to the camera view limited by the robot’s height and the narrow Field of View (FOV) of the inexpensive camera. In this research, auto-parking of a small-scale mobile robot with a front camera only was achieved in a four-step process: Firstly, transfer learning was performed on the AlexNet, a popular pre-trained convolutional neural network (CNN). It was trained with 150 pictures of empty parking slots and 150 pictures of occupied parking slots from the view angle of a small-scale robot. The dataset of images was divided into a group of 70% images for training and the remaining 30% images for validation. An average success rate of 95% was achieved. Secondly, the image of detected empty parking space was processed with edge detection followed by the computation of parametric representations of the boundary lines using the Hough Transform algorithm. Thirdly, the positions of the entrance point and center of available parking space were predicted based on the robot kinematic model as the robot was driving closer to the parking space because the boundary lines disappeared partially or completely from its camera view due to the height and FOV limitations. The robot used its wheel speeds to compute the positions of the parking space with respect to its changing local frame as it moved along, based on its kinematic model. Lastly, the predicted entrance point of the parking space was used as the reference for the motion control of the robot until it was replaced by the actual center when it became visible again by the robot. The linear and angular velocities of the robot chassis center were computed based on the error between the current chassis center and the reference point. Then the left and right wheel speeds were obtained using inverse kinematics and sent to the motor driver. The above-mentioned four subtasks were all successfully accomplished, with the transformed learning, image processing, and target prediction performed in MATLAB, while the motion control and image capture conducted on a self-built small scale differential drive mobile robot. The small-scale robot employs a Raspberry Pi board, a Pi camera, an L298N dual H-bridge motor driver, a USB power module, a power bank, four wheels, and a chassis. Future research includes three areas: the integration of all four subsystems into one hardware/software platform with the upgrade to an Nvidia Jetson Nano board that provides superior performance for deep learning and image processing; more testing and validation on the identification of available parking space and its boundary lines; improvement of performance after the hardware/software integration is completed.

Keywords: autonomous parking, convolutional neural network, image processing, kinematics-based prediction, transfer learning

Procedia PDF Downloads 133