Boosted Convolutional Neural Network for object recognition at large scale

Brahimi, Sourour; Aoun, Najib Ben; Amar, Chokri Ben

doi:10.1016/j.neucom.2018.11.031

Cited by 12 publications

(4 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Although the VGG16 network increases the extraction ability of detailed features by the combination and stacking of 3 × 3 filters, the distinguishing ability of features between smoke and smoke-like images such as cloud and fog is insufficiently strong. To further strengthen the significance of the features, this paper draws on the convolutional layer enhancement idea in literature [18], for cascading two convolutional layers after each convolutional layer of the VGG16 network.…”

Section: Our Methodsmentioning

confidence: 99%

Video Smoke Detection Method Based on Change-Cumulative Image and Fusion Deep Network

Liu

Cheng

et al. 2019

Sensors

View full text Add to dashboard Cite

Smoke detection technology based on computer vision is a popular research direction in fire detection. This technology is widely used in outdoor fire detection fields (e.g., forest fire detection). Smoke detection is often based on features such as color, shape, texture, and motion to distinguish between smoke and non-smoke objects. However, the salience and robustness of these features are insufficiently strong, resulting in low smoke detection performance under complex environment. Deep learning technology has improved smoke detection performance to a certain degree, but extracting smoke detail features is difficult when the number of network layers is small. With no effective use of smoke motion characteristics, indicators such as false alarm rate are high in video smoke detection. To enhance the detection performance of smoke objects in videos, this paper proposes a concept of change-cumulative image by converting the YUV color space of multi-frame video images into a change-cumulative image, which can represent the motion and color-change characteristics of smoke. Then, a fusion deep network is designed, which increases the depth of the VGG16 network by arranging two convolutional layers after each of its convolutional layer. The VGG16 and Resnet50 (Deep residual network) network models are also arranged using the fusion deep network to improve feature expression ability while increasing the depth of the whole network. Doing so can help extract additional discriminating characteristics of smoke. Experimental results show that by using the change-cumulative image as the input image of the deep network model, smoke detection performance is superior to the classic RGB input image; the smoke detection performance of the fusion deep network model is better than that of the single VGG16 and Resnet50 network models; the smoke detection accuracy, false positive rate, and false alarm rate of this method are better than those of the current popular methods of video smoke detection.

show abstract

Section: Our Methodsmentioning

confidence: 99%

Video Smoke Detection Method Based on Change-Cumulative Image and Fusion Deep Network

Liu

Cheng

et al. 2019

Sensors

View full text Add to dashboard Cite

show abstract

“…In the field of image processing, there are many literatures on object recognition [21][22][23][24][25]. As can be seen from Figure 4, the type of complex structure is mainly reflected by its projection shape on the Y Z plane.…”

Section: Complex Structure Recognitionmentioning

confidence: 99%

Power Pylon Reconstruction Based on Abstract Template Structures Using Airborne LiDAR Data

et al. 2019

View full text Add to dashboard Cite

As an important power facility for transmission corridors, automatic three-dimensional (3D) reconstruction of the pylon plays an important role in the development of smart grid. In this study, a novel three-dimensional reconstruction method using airborne LiDAR (Light Detection And Ranging) point cloud is developed and tested. First, a principal component analysis (PCA) algorithm is performed for pylon redirection based on the structural feature of the upper part of a pylon. Then, based on the structural similarity of a pylon, a pylon is divided into three parts that are inverted triangular pyramid lower structures, quadrangular frustum pyramid middle structures, and complex upper or lateral structures. The reconstruction of the inverted triangular pyramid structures and quadrangular frustum pyramid structures is based on prior knowledge and a data-driven strategy, where the 2D alpha shape algorithm is used to obtain contour points and 2D linear fitting is carried out based on the random sample consensus (RANSAC) method. Complex structures’ reconstruction is based on the priori abstract template structure and a data-driven strategy, where the abstract template structure is used to determine the topological relationship among corner points and the image processing method is used to extract corner points of the abstract template structure. The main advantages in the proposed method include: (1) Improving the accuracy of the pylon decomposition method through introducing a new feature to identify segmentation positions; (2) performing the internal structure of quadrangular frustum pyramids reconstruction; (3) establishing the abstract template structure and using image processing methods to improve computational efficiency of pylon reconstruction. Eight types of pylons are tested in this study, and the average error of pylon reconstruction is 0.32 m and the average of computational time is 0.8 s. These results provide evidence that the pylon reconstruction method developed in this study has high accuracy, efficiency, and applicability.

show abstract

“…Ahn and Chung in [22] combined the advantages from the multiscale Harris corner and SIFT descriptors to make an object detection using depth range features for metric information. Zheng, Barahimi, Aoun and Amar [23] present a boosted convolutional neural network for object recognition, using a boosted blocks in a succession of convolutional layers. Pertusa, Gallego and Bernabeu [24] propose an application for smartphones that allows object recognition using CNN.…”

Section: Introductionmentioning

confidence: 99%

Indoor Localization and Navigation based on Deep Learning using a Monocular Visual System

Ancona¹,

Corona-Ramírez²,

Frías³

2021

IJACSA

View full text Add to dashboard Cite

Now-a-days, computer systems are important for artificial vision systems to analyze the acquired data to realize crucial tasks, such as localization and navigation. For successful navigation, the robot must interpret the acquired data and determine its position to decide how to move through the environment. This paper proposes an indoor mobile robot visuallocalization and navigation approach for autonomous navigation. A convolutional neural network and background modeling are used to locate the system in the environment. Object detection is based on copy-move detection, an image forensic technique, extracting features from the image to identify similar regions. An adaptive threshold is proposed due to the illumination changes. The detected object is classified to evade it using a control deep neural network. A U-Net model is implemented to track the path trajectory. The experiment results were obtained from real data, proving the efficiency of the proposed algorithm. The adaptive threshold solves illumination variation issues for object detection.

show abstract

Boosted Convolutional Neural Network for object recognition at large scale

Cited by 12 publications

References 24 publications

Video Smoke Detection Method Based on Change-Cumulative Image and Fusion Deep Network

Video Smoke Detection Method Based on Change-Cumulative Image and Fusion Deep Network

Power Pylon Reconstruction Based on Abstract Template Structures Using Airborne LiDAR Data

Indoor Localization and Navigation based on Deep Learning using a Monocular Visual System

Contact Info

Product

Resources

About