scholarly journals A Two-Stream CNN Model with Adaptive Adjustment of Receptive Field Dedicated to Flame Region Detection

Symmetry ◽  
2021 ◽  
Vol 13 (3) ◽  
pp. 397
Author(s):  
Peng Lu ◽  
Yaqin Zhao ◽  
Yuan Xu
Keyword(s):  
Receptive Field ◽  
State Of The Art ◽  
Region Of Interest ◽  
Surveillance Systems ◽  
Dynamic Features ◽  
Region Detection ◽  
Significant Performance ◽  
Hardware Configuration ◽  
Flame Region ◽  
Multiple Metrics

Convolutional neural networks (CNN) have yielded state-of-the-art performance in image segmentation. Their application in video surveillance systems can provide very useful information for extinguishing fire in time. The current studies mostly focused on CNN-based flame image classification and have achieved good accuracy. However, the research of CNN-based flame region detection is extremely scarce due to the bulky network structures and high hardware configuration requirements of the state-of-the-art CNN models. Therefore, this paper presents a two-stream convolutional neural network for flame region detection (TSCNNFlame). TSCNNFlame is a lightweight CNN architecture including a spatial stream and temporal stream for detecting flame pixels in video sequences captured by fixed cameras. The static features from the spatial stream and dynamic features from the temporal stream are fused by three convolutional layers to reduce the false positives. We replace the convolutional layer of CNN with the selective kernel (SK)-Shuffle block constructed by integrating the SK convolution into the deep convolutional layer of ShuffleNet V2. The SKnet blocks can adaptively adjust the size of one receptive field with the proportion of one region of interest (ROI) in it. The grouped convolution used in Shufflenet solves the problem in which the multi-branch structure of SKnet causes the network parameters to double with the number of branches. Therefore, the CNN network dedicated to flame region detection balances the efficiency and accuracy by the lightweight architecture, the temporal–spatial features fusion, and the advantages of the SK-Shuffle block. The experimental results, which are evaluated by multiple metrics and are analyzed from many angles, show that this method can achieve significant performance while reducing the running time.

Experimental Validation of Minimax Entropy Principle in Ultrasound Images

2019 ◽  
Vol 12 (6) ◽  
pp. 487-493
Author(s):  
Neha Mehta ◽  
Svav Prasad ◽  
Leena Arya
Keyword(s):  
Region Of Interest ◽  
Ultrasound Images ◽  
Human Gallbladder ◽  
New Approach ◽  
Non Invasive ◽  
Entropy Principle ◽  
Significant Performance ◽  
Considerable Impact ◽  
Ultrasonic Devices ◽  
Measure Of Uncertainty

Ultrasound imaging is one of the non-invasive imaging, that diagnoses the disease inside a human body and there are numerous ultrasonic devices being used frequently. Entropy as a well known statistical measure of uncertainty has a considerable impact on the medical images. A procedure for minimizing the entropy with respect to the region of interest is demonstrated. This new approach has shown the experiments using Extracted Region Of Interest Based Sharpened image, called as (EROIS) image based on Minimax entropy principle and various filters. In this turn, the approach also validates the versatility of the entropy concept. Experiments have been performed practically on the real-time ultrasound images collected from ultrasound centers and have shown a significant performance. The present approach has been validated with showing results over ultrasound images of the Human Gallbladder.

scholarly journals A Collaborative Region Detection and Grading Framework for Forest Fire Smoke Using Weakly Supervised Fine Segmentation and Lightweight Faster-RCNN

Forests ◽  
2021 ◽  
Vol 12 (6) ◽  
pp. 768
Author(s):  
Jin Pan ◽  
Xiaoming Ou ◽  
Liang Xu
Keyword(s):  
Forest Fire ◽  
Forest Fires ◽  
Learning Strategy ◽  
Early Stage ◽  
Fire Detection ◽  
Surveillance Systems ◽  
Region Detection ◽  
Decision Network ◽  
Fire Smoke ◽  
Weakly Supervised

Forest fires are serious disasters that affect countries all over the world. With the progress of image processing, numerous image-based surveillance systems for fires have been installed in forests. The rapid and accurate detection and grading of fire smoke can provide useful information, which helps humans to quickly control and reduce forest losses. Currently, convolutional neural networks (CNN) have yielded excellent performance in image recognition. Previous studies mostly paid attention to CNN-based image classification for fire detection. However, the research of CNN-based region detection and grading of fire is extremely scarce due to a challenging task which locates and segments fire regions using image-level annotations instead of inaccessible pixel-level labels. This paper presents a novel collaborative region detection and grading framework for fire smoke using a weakly supervised fine segmentation and a lightweight Faster R-CNN. The multi-task framework can simultaneously implement the early-stage alarm, region detection, classification, and grading of fire smoke. To provide an accurate segmentation on image-level, we propose the weakly supervised fine segmentation method, which consists of a segmentation network and a decision network. We aggregate image-level information, instead of expensive pixel-level labels, from all training images into the segmentation network, which simultaneously locates and segments fire smoke regions. To train the segmentation network using only image-level annotations, we propose a two-stage weakly supervised learning strategy, in which a novel weakly supervised loss is proposed to roughly detect the region of fire smoke, and a new region-refining segmentation algorithm is further used to accurately identify this region. The decision network incorporating a residual spatial attention module is utilized to predict the category of forest fire smoke. To reduce the complexity of the Faster R-CNN, we first introduced a knowledge distillation technique to compress the structure of this model. To grade forest fire smoke, we used a 3-input/1-output fuzzy system to evaluate the severity level. We evaluated the proposed approach using a developed fire smoke dataset, which included five different scenes varying by the fire smoke level. The proposed method exhibited competitive performance compared to state-of-the-art methods.

scholarly journals Plant Leaf Disease Recognition Using Depth-Wise Separable Convolution-Based Models

Symmetry ◽  
2021 ◽  
Vol 13 (3) ◽  
pp. 511
Author(s):  
Syed Mohammad Minhaz Hossain ◽  
Kaushik Deb ◽  
Pranab Kumar Dhar ◽  
Takeshi Koshiba
Keyword(s):  
State Of The Art ◽  
Computational Cost ◽  
Region Of Interest ◽  
Number Of Clusters ◽  
Plant Leaf ◽  
Leaf Disease ◽  
Automatic Initialization ◽  
Adequate Accuracy ◽  
Model Size ◽  
High Computational Cost

Proper plant leaf disease (PLD) detection is challenging in complex backgrounds and under different capture conditions. For this reason, initially, modified adaptive centroid-based segmentation (ACS) is used to trace the proper region of interest (ROI). Automatic initialization of the number of clusters (K) using modified ACS before recognition increases tracing ROI’s scalability even for symmetrical features in various plants. Besides, convolutional neural network (CNN)-based PLD recognition models achieve adequate accuracy to some extent. However, memory requirements (large-scaled parameters) and the high computational cost of CNN-based PLD models are burning issues for the memory restricted mobile and IoT-based devices. Therefore, after tracing ROIs, three proposed depth-wise separable convolutional PLD (DSCPLD) models, such as segmented modified DSCPLD (S-modified MobileNet), segmented reduced DSCPLD (S-reduced MobileNet), and segmented extended DSCPLD (S-extended MobileNet), are utilized to represent the constructive trade-off among accuracy, model size, and computational latency. Moreover, we have compared our proposed DSCPLD recognition models with state-of-the-art models, such as MobileNet, VGG16, VGG19, and AlexNet. Among segmented-based DSCPLD models, S-modified MobileNet achieves the best accuracy of 99.55% and F1-sore of 97.07%. Besides, we have simulated our DSCPLD models using both full plant leaf images and segmented plant leaf images and conclude that, after using modified ACS, all models increase their accuracy and F1-score. Furthermore, a new plant leaf dataset containing 6580 images of eight plants was used to experiment with several depth-wise separable convolution models.

scholarly journals High Order Volumetric Directional Pattern for Video-Based Face Recognition

2019 ◽  
Vol 2019 ◽  
pp. 1-10 ◽  
Author(s):  
Almabrok Essa ◽  
Vijayan Asari
Keyword(s):  
State Of The Art ◽  
Directional Pattern ◽  
High Order ◽  
Dynamic Textures ◽  
Dynamic Features ◽  
Novel Approach ◽  
Variation Patterns ◽  
Directional Variation ◽  
Video Based Face Recognition ◽  
Information Dynamic

Describing the dynamic textures has attracted growing attention in the field of computer vision and pattern recognition. In this paper, a novel approach for recognizing dynamic textures, namely, high order volumetric directional pattern (HOVDP), is proposed. It is an extension of the volumetric directional pattern (VDP) which extracts and fuses the temporal information (dynamic features) from three consecutive frames. HOVDP combines the movement and appearance features together considering the nth order volumetric directional variation patterns of all neighboring pixels from three consecutive frames. In experiments with two challenging video face databases, YouTube Celebrities and Honda/UCSD, HOVDP clearly outperformed a set of state-of-the-art approaches.

scholarly journals Fast horizon detection in maritime images using region-of-interest

2018 ◽  
Vol 14 (7) ◽  
pp. 155014771879075 ◽  
Author(s):  
Chi Yoon Jeong ◽  
Hyun S Yang ◽  
KyeongDeok Moon
Keyword(s):  
Edge Detection ◽  
State Of The Art ◽  
Region Of Interest ◽  
Least Square ◽  
Fast Method ◽  
Multi Scale ◽  
Art Methods ◽  
Horizon Line ◽  
Horizon Detection ◽  
Interest Detection

In this article, we propose a fast method for detecting the horizon line in maritime scenarios by combining a multi-scale approach and region-of-interest detection. Recently, several methods that adopt a multi-scale approach have been proposed, because edge detection at a single is insufficient to detect all edges of various sizes. However, these methods suffer from high processing times, requiring tens of seconds to complete horizon detection. Moreover, the resolution of images captured from cameras mounted on vessels is increasing, which reduces processing speed. Using the region-of-interest is an efficient way of reducing the amount of processing information required. Thus, we explore a way to efficiently use the region-of-interest for horizon detection. The proposed method first detects the region-of-interest using a property of maritime scenes and then multi-scale edge detection is performed for edge extraction at each scale. The results are then combined to produce a single edge map. Then, Hough transform and a least-square method are sequentially used to estimate the horizon line accurately. We compared the performance of the proposed method with state-of-the-art methods using two publicly available databases, namely, Singapore Marine Dataset and buoy dataset. Experimental results show that the proposed method for region-of-interest detection reduces the processing time of horizon detection, and the accuracy with which the proposed method can identify the horizon is superior to that of state-of-the-art methods.

scholarly journals Image Dehazing Based on (CMTnet) Cascaded Multi-scale Convolutional Neural Networks and Efficient Light Estimation Algorithm

2020 ◽  
Vol 10 (3) ◽  
pp. 1190
Author(s):  
Samia Haouassi ◽  
Di Wu
Keyword(s):  
State Of The Art ◽  
Estimation Algorithm ◽  
Surveillance Systems ◽  
Security Systems ◽  
High Quality ◽  
Image Dehazing ◽  
Multi Scale ◽  
Second Stage ◽  
Computer Vision Applications ◽  
Atmospheric Light

Image dehazing plays a pivotal role in numerous computer vision applications such as object recognition, surveillance systems, and security systems, where it can be considered as an introductory stage. Recently, many proposed learning-based works address this significant task; however, most of them neglect the atmospheric light estimation and fail to produce accurate transmission maps. To address such a problem, in this paper, we propose a two-stage dehazing system. The first stage presents an accurate atmospheric light algorithm labeled “A-Est” that employs hazy image blurriness and quadtree decomposition. Te second stage represents a cascaded multi-scale CNN model called CMT n e t that consists of two subnetworks, one for calculating rough transmission maps (CMCNN t r ) and the other for its refinement (CMCNN t ). Each subnetwork is composed of three-layer D-units (D indicates dense). Experimental analysis and comparisons with state-of-the-art dehazing methods revealed that the proposed system can estimate AL and t efficiently and accurately by achieving high-quality dehazing results and outperforms state-of-the-art comparative methods according to SSIM and MSE values, where our proposed achieves the best scores of both (91% average SSIM and 0.068 average MSE).

iHDT++: improving HDT for SPARQL triple pattern resolution

2020 ◽  
Vol 39 (2) ◽  
pp. 2249-2261
Author(s):  
Antonio Hernández-Illera ◽  
Miguel A. Martínez-Prieto ◽  
Javier D. Fernández ◽  
Antonio Fariña
Keyword(s):  
State Of The Art ◽  
Simple Design ◽  
Performance Improvements ◽  
Semantic Data ◽  
Object Based ◽  
Efficient Access ◽  
Memory Footprint ◽  
Significant Performance ◽  
Pattern Resolution

RDF self-indexes compress the RDF collection and provide efficient access to the data without a previous decompression (via the so-called SPARQL triple patterns). HDT is one of the reference solutions in this scenario, with several applications to lower the barrier of both publication and consumption of Big Semantic Data. However, the simple design of HDT takes a compromise position between compression effectiveness and retrieval speed. In particular, it supports scan and subject-based queries, but it requires additional indexes to resolve predicate and object-based SPARQL triple patterns. A recent variant, HDT++, improves HDT compression ratios, but it does not retain the original HDT retrieval capabilities. In this article, we extend HDT++ with additional indexes to support full SPARQL triple pattern resolution with a lower memory footprint than the original indexed HDT (called HDT-FoQ). Our evaluation shows that the resultant structure, iHDT++ , requires 70 - 85% of the original HDT-FoQ space (and up to 48 - 72% for an HDT Community variant). In addition, iHDT++ shows significant performance improvements (up to one level of magnitude) for most triple pattern queries, being competitive with state-of-the-art RDF self-indexes.

Remote Monitoring System of Robotic Car Based on Internet of Things Using Raspberry Pi

2020 ◽  
Vol 17 (5) ◽  
pp. 2288-2295
Author(s):  
K. V. Sowmya ◽  
Harshavardhan Jamedar ◽  
Pradeep Godavarthi
Keyword(s):  
User Interface ◽  
Monitoring System ◽  
Remote Monitoring ◽  
Fire Risk ◽  
Third Party ◽  
Raspberry Pi ◽  
Surveillance Systems ◽  
Dynamic Features ◽  
Remote Monitoring System ◽  
Security Officers

Development of surveillance and monitoring systems are quite difficult and challenging task at times. The design of a system depends on the environment to be monitored. Such surveillance systems need to have dynamic features, for e.g., cameras used for monitoring may be mobiles, web cams etc. installed to the system. Such systems are used in various large buildings like shopping malls where it could incur high cost for installing cameras in each level of buildings. Even for people like security officers it could be huge task to cover an entire building. Other examples for dynamic surveillance system could be detecting poisonous gases in an area, explosives and any fire risk elements. Another case is that it can reach where the area is not accessed by humans. In view of these challenges we propose a Remote monitoring system where a Robotic Car is installed with camera, Ultrasonic sensor, DHT11, PIR sensors according to the environment involved. The instructions are given to the robotic car using a third party app called Blynk as user interface. Here the raspberry pi is used as a microcontroller which is connected to WIFI acts as the communication medium to connect the server provided by Blynk. The Blynk app which acts as a user interface is interacted with the car using Wi-Fi and its server.

Sign in / Sign up

Export Citation Format

Share Document