Appropriate CNN Architecture and Optimizer for Vehicle Type Classification System on the Toll Road

Swastika, Windra; Ariyanto, Martien Febriant; Setiawan, Hendry; Irawan, Paulus Lucky Tirma

doi:10.1088/1742-6596/1196/1/012044

Cited by 9 publications

(4 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Each optimizer is evaluated with three different CNN architectures such as LeNet, shallow network, and MiniVGGNet. The outcome illustrates that Adadelta optimizer is considered to be best for the Mini VGGNet architecture due to high accuracy optimizers [21].…”

Section: Literature Reviewmentioning

confidence: 98%

Classification of fruits ripeness using CNN with multivariate analysis by SGD

Sumathi¹,

Vinod²

2022

NNW

View full text Add to dashboard Cite

Ripeness estimation of fruits is an essential process that impact the quality of fruits and its marketing. Nearly 30% to 35% get wasted from the harvested fruits due to lack of skilled workers in classification and fruit grading. Although it can be executed by human assessment, it is time consuming, costlier and error prone. Lot of research is carried to automate the quality assessment of fruits. Several hyper-parameters have been considered which have liven up by providing robust convolutional neural network (CNN). This paper has focused on image resizer stochastic gradient descent (SGD) algorithm for computing the loss. It updates the parameter by concentrating channels with respect to red, green, and blue (RGB) to identify and classify the images as ripen and rotten. The real time dataset (6702 images) of oranges, papaya and banana is collected. Using SGD optimizer, learning rate of 0.01 and nearest neighbor interpolation algorithm as resizer, the proposed model has achieved accuracy rate of 96.56% after 38 epochs in classifying the fruits as ripen and rotten. It is also observed that it is possible to use small dataset on visual geometry group with 16 layer (VGG) with the above specification and good accuracy rate can be achieved.

show abstract

Section: Literature Reviewmentioning

confidence: 98%

Classification of fruits ripeness using CNN with multivariate analysis by SGD

Sumathi¹,

Vinod²

2022

NNW

View full text Add to dashboard Cite

show abstract

“…Penelitian sebelumnya sudah melakukan percobaan dengan menggunakan metode CNN dan hasil yang diperoleh dari jaringan memiliki evaluasi yang akurat yaitu 73%. Percobaan tersebut menggunakan arsitektur MiniVGGNet yang menerapkan fungsi optimasi adadelta dan parameter citra 64x64 dan epoch 40 [5]. Existing CNN yang akan digunakan dalam penelitian ini antara lain adalah Alexnet dan Mobilenet V2 yang telah berhasil mengklasifikasikan objek dengan baik.…”

Section: Pendahuluanunclassified

Analisis Klasifikasi Mobil Pada Gardu Tol Otomatis (GTO) Menggunakan Convolutional Neural Network (CNN)

et al. 2022

View full text Add to dashboard Cite

The concept of a smart city is the most important issue in the development aspect of big cities in the world. Where the city must promise a more comfortable, organized, healthy and efficient life. Smart transportation is part of a smart city that is useful for improving better urban planning. Smart transportation also applies to toll roads, such as automating toll road retribution payments. Automatic Toll Gate (GTO) in Indonesia still uses sensors. However, sensors often misclassify trailers. In addition, the use of sensors also requires additional costs in installation and maintenance. Currently, every toll gate is equipped with cameras for various purposes. By utilizing the camera for vehicle type classification, the cost of the GTO will be reduced. For this reason, utilizing a digital camera with computer vision for vehicle type classification is the solution. Convolutional Neural Networks (CNN) is the most popular technique today in solving computer vision problems. Exploit the existing CNN by replacing the last fully connected output according to the number of vehicle classes. The test results show that mobilenet V2 is better in the classification of vehicle types, the best accuracy is Alexnet 93.81% and Mobilenet 96.19%. Computer vision by utilizing CNN is expected to replace the use of sensors so that implementation costs are cheaper.

show abstract

“…RMSProp was introduced to address the problem of the monotonically decreasing learning rate [23]. The weights are updated based on Equation (17),…”

Section: Rmsprop Optimizermentioning

confidence: 99%

“…For this architecture, RMSProp achieved the best accuracy. Swastika et al [ 17 ] evaluated three optimizers to classify vehicle types: Adam, Adadelta, and SGD. The authors used three CNN architectures to evaluate each optimizer: a shallow network, LeNet, and MiniVGGNet.…”

Section: Related Workmentioning

confidence: 99%

Comparative Study of First Order Optimizers for Image Classification Using Convolutional Neural Networks on Histopathology Images

2020

View full text Add to dashboard Cite

The classification of histopathology images requires an experienced physician with years of experience to classify the histopathology images accurately. In this study, an algorithm was developed to assist physicians in classifying histopathology images; the algorithm receives the histopathology image as an input and produces the percentage of cancer presence. The primary classifier used in this algorithm is the convolutional neural network, which is a state-of-the-art classifier used in image classification as it can classify images without relying on the manual selection of features from each image. The main aim of this research is to improve the robustness of the classifier used by comparing six different first-order stochastic gradient-based optimizers to select the best for this particular dataset. The dataset used to train the classifier is the PatchCamelyon public dataset, which consists of 220,025 images to train the classifier; the dataset is composed of 60% positive images and 40% negative images, and 57,458 images to test its performance. The classifier was trained on 80% of the images and validated on the rest of 20% of the images; then, it was tested on the test set. The optimizers were evaluated based on their AUC of the ROC curve. The results show that the adaptative based optimizers achieved the highest results except for AdaGrad that achieved the lowest results.

show abstract

Appropriate CNN Architecture and Optimizer for Vehicle Type Classification System on the Toll Road

Cited by 9 publications

References 2 publications

Classification of fruits ripeness using CNN with multivariate analysis by SGD

Classification of fruits ripeness using CNN with multivariate analysis by SGD

Analisis Klasifikasi Mobil Pada Gardu Tol Otomatis (GTO) Menggunakan Convolutional Neural Network (CNN)

Comparative Study of First Order Optimizers for Image Classification Using Convolutional Neural Networks on Histopathology Images

Contact Info

Product

Resources

About