Inclined Image Recognition for Aerial Mapping using Deep Learning and Tree based Models

Attamimi, Muhammad; Mardiyanto, Ronny; Irfansyah, Astria Nur

doi:10.12928/telkomnika.v16i6.10157

Cited by 6 publications

(6 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Figure 2 shows the illustration of our proposed method. The mechanism of transfer learning is similar to our previous work in [33] which is by collecting datasets consist of seven types of facial expressions and training the publicly available model with the new dataset. Since model building is time consuming, the training phase is conducted offline.…”

Section: Proposed Methodsmentioning

confidence: 99%

Deep learning based facial expressions recognition system for assisting visually impaired persons

2020

View full text Add to dashboard Cite

In general, a good interaction including communication can be achieved when verbal and non-verbal information such as body movements, gestures, facial expressions, can be processed in two directions between the speaker and listener. Especially the facial expression is one of the indicators of the inner state of the speaker and/or the listener during the communication. Therefore, recognizing the facial expressions is necessary and becomes the important ability in communication. Such ability will be a challenge for the visually impaired persons. This fact motivated us to develop a facial recognition system. Our system is based on deep learning algorithm. We implemented the proposed system on a wearable device which enables the visually impaired persons to recognize facial expressions during the communication. We have conducted several experiments involving the visually impaired persons to validate our proposed system and the promising results were achieved.

show abstract

Section: Proposed Methodsmentioning

confidence: 99%

Deep learning based facial expressions recognition system for assisting visually impaired persons

2020

View full text Add to dashboard Cite

show abstract

“…The designed system focuses on detecting handguns in minimum training time with high accuracy results. A pre-trained model such as GoogleNet, VGGNet-19 and MobileNet have been trained with more than million images to minimize the object detection errors in images [19][20][21]. Based on fast and accurately training properties, a MobileNetv3 model is used in this method to classify and detect handguns accurately.…”

Section: The Proposed Modelmentioning

confidence: 99%

The detection of handguns from live-video in real-time based on deep learning

2020

View full text Add to dashboard Cite

Many people have been killed indiscriminately by the use of handguns in different countries. Terrorist acts, online fighting games and mentally disturbed people are considered the common reasons for these crimes. A realtime handguns detection surveillance system is built to overcome these bad acts, based on convolutional neural networks (CNNs). This method is focused on the detection of different weapons, such as (handgun and rifles). The identification of handguns from surveillance cameras and images requires monitoring by human supervisor, that can cause errors. To overcome this issue, the designed detection system sends an alert message to the supervisor when a weapon is detected. In the proposed detection system, a pre-trained deep learning model MobileNetV3-SSDLite is used to perform the handgun detection operation. This model has been selected because it is fast and accurate in infering to integrate network for detecting and classifying weapons in images. The experimental result using global handguns datasets of various weapons showed that the use of MobileNetV3 with SSDLite model both enhance the accuracy level in identifying the real time handguns detection.

show abstract

“…The autocorrelation function carries sufficiently complete information about the character of the input image [1], [2], [3], [4], [5] in addition, it is invariant towards to the description of moving images in the vertical and horizontal directions.…”

Section: Recognition Systems Problemsmentioning

confidence: 99%

“…The sphere with the center at the point, corresponding to the image S, formed by the radius r, will be called r -region of the image S (Fig. 3); Thus, for a rather large meaning k, this rule defines the number of the representation of the images 1 V that are contained within a sphere with a radius r centered at the point, corresponding to the image S. This number is compared with the number of representation 2 V of the images in the same sphere. With this method, solutions are found that are almost unintelligible to the errors that arise due to the effects of interference.…”

Section: Let the Given Imagementioning

confidence: 99%

Обробка Інформації Оптичним Датчиком Робота За Умови Неповної Вхідної Інформації

Parkhomey

Batrak

Tsopa

et al. 2019

«Адаптивні Системи Автоматичного Управління Міжвідомчий науково

View full text Add to dashboard Cite

The range of tasks, solved by the properties of classification and generalization, is quite broad. However, the quality of solving to the problem, for one reason or another, is not always the same. Some tasks cannot be solved at once, but a detailed study of phenomenon allows us to make the necessary classification. For instance, the task is to divide the objects, with the same geometric dimensions and objects, by visual observation into two classes. Moreover, the part is madeof steel, and another one -of gypsum. In principle, such task cannot be solved.To classify these subjects, a more detailed study of them is required. It is sufficiently to get data on the weight of objects, so that the task of classification has become trivial. The difficulty is to find such a description of phenomenon, in which the image (phenomenon), within each class, would have defined similar properties. Therefore, in order to eliminate the effect of interference on the recognition process, it is proposed to introduce a special decisive rule based on the fact that the decision to attach the image to the image is made on the basis of an analysis of images that have fallen into a definite close space that is classified.

show abstract

Inclined Image Recognition for Aerial Mapping using Deep Learning and Tree based Models

Cited by 6 publications

References 8 publications

Deep learning based facial expressions recognition system for assisting visually impaired persons

Deep learning based facial expressions recognition system for assisting visually impaired persons

The detection of handguns from live-video in real-time based on deep learning

Обробка Інформації Оптичним Датчиком Робота За Умови Неповної Вхідної Інформації

Contact Info

Product

Resources

About