Machine learning-based ship detection and tracking using satellite images for maritime surveillance

Wang, Yu; Rajesh, G.; Raajini, X. Mercilin; Kritika, N.; Kavinkumar, A.; Shah, Syed Bilal Hussain

doi:10.3233/ais-210610

Cited by 38 publications

(9 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Wang Yu et al [3] developed a model to detect the images of ship in high-resolution remote sensing images. SDT filter is used to perform the pre-processing, which reduces the noise.…”

Section: Methodsmentioning

confidence: 99%

Hash-Based Deep Learning Approach for Remote Sensing Satellite Imagery Detection

Gadamsetty

Rupa

et al. 2022

Water

View full text Add to dashboard Cite

Ship detection plays a crucial role in marine security in remote sensing imagery. This paper discusses about a deep learning approach to detect the ships from satellite imagery. The model developed in this work achieves integrity by the inclusion of hashing. This model employs a supervised image classification technique to classify images, followed by object detection using You Only Look Once version 3 (YOLOv3) to extract features from deep CNN. Semantic segmentation and image segmentation is done to identify object category of each pixel using class labels. Then, the concept of hashing using SHA-256 is applied in conjunction with the ship count and location of bounding box in satellite image. The proposed model is tested on a Kaggle Ships dataset, which consists of 231,722 images. A total of 70% of this data is used for training, and the 30% is used for testing. To add security to images with detected ships, the model is enhanced by hashing using SHA-256 algorithm. Using SHA-256, which is a one-way hash, the data are split up into blocks of 64 bytes. The input data to the hash function are both the ship count and bounding box location. The proposed model achieves integrity by using SHA-256. This model allows secure transmission of highly confidential images that are tamper-proof.

show abstract

“…Wang Yu et al [3] developed a model to detect the images of ship in high-resolution remote sensing images. SDT filter is used to perform the pre-processing, which reduces the noise.…”

Section: Methodsmentioning

confidence: 99%

Hash-Based Deep Learning Approach for Remote Sensing Satellite Imagery Detection

Gadamsetty

Rupa

et al. 2022

Water

View full text Add to dashboard Cite

show abstract

“…Machine learning techniques can improve wireless network security, reduce congestion, and aid in physical layer authentication and error detection [17][18][19]. ML techniques also help analyze WSN packets and identify problematic nodes [26][27][28][29].…”

Section: Literature Surveymentioning

confidence: 99%

A Novel Model for Network Intrusion Detection by Using Statistical Learning Approach

Jeevaraj¹,

B.²,

M.³

et al. 2022

INDJCSE

View full text Add to dashboard Cite

Network intrusion detection is the pressing need of every communication network. Network Intrusion Detection Systems (NIDSes) play an important role in security operations to detect and defend against cyberattacks. As artificial intelligence (AI)-powered NIDSes are adaptive to various kinds of attacks by exploring the knowledge presented in the data, they are in high demand to treat the cyber-attacks nowadays with increasing diversity and intensity. This research works finds that the efficient learning to overcome these issues. The NBMU model performs well and it holds an accuracy 89.44%. The NBM shows worst output which as 65.80%. The NBMU and NB models perform well and it have same precision value 0.89. The NBM model has poor performance which as 0.66. The NBMU and NB models perform well and it have same recall value 0.89. The NBM model has poor performance which as 0.66. The NBMU and NB models perform well and it have same F-Measure value 0.89. The NBM model has poor performance which as 0.64. The NBMU has ROC level 0.90 which as highest performance compare with other models. The NBM model has shown the worst outcome which as 0.59.The NBMU has highest PRC level 0.90. The NBM model has not satisfactory which as 0.59 of PRC value. The NBMU and NB has the highest MCC 0.88. The NBM model has poor performance which as 0.36 of MCC value. The NBMU and NB has same kappa 0.88. The NBM model has poor performance which as 0.32 of kappa value. The NBMU and NB has the least MAE 0.09. The NBM model has poor performance which as 0.32 of MAE value. The NBM has poor performance which as 0.89.The NB has the least RAE 10.52%. The NBU modes has poor performance as well same RAE which as 116.64%. The NBU modes has poor performance as well same RRSE which as 132.51%.The NBMU has poor performance for making its model which as 20.02 seconds.

show abstract

“…[18][19][20] The main concept is to put IR array sensors and monitor the temperature pattern of the room in search of rapid changes caused by a person's presence. [21] The use of RGB cameras is a relatively common method for tracking and counting people [22]. WiFi signals have been employed for both sensing and communication.…”

Section: Literature Surveymentioning

confidence: 99%

A Second Order Inductive Deduction Approach for Identification of Room Occupancy by Wireless Sensors

Jeevaraj¹,

B.²,

M.³

et al. 2022

INDJCSE

View full text Add to dashboard Cite

Occupancy detection is crucial in many smart building applications, including reducing building energy consumption by managing heating, ventilation, and air conditioning systems, monitoring systems, and lighting system management, tracking patients in hospitals for medical issues, advertising to shoppers in malls, and search and rescue missions. The global positioning system is most frequently employed as a localization technique, yet it is incredibly imprecise when used indoors. The interior environment is challenging to manage because, in addition to the signal loss, privacy is a significant issue. Indoor tracking and wireless sensor network sensor localization share many similarities. Machine Learning helps to overcome the mentioned issues. This research works finds that the Attribute Selected Classifier with Naïve Bayes Updateable of second order ensemble model gives highest performance which as accuracy level 86.69%%, kappa statistic value 0.68, precision value 0.87, recall value 0.87, F-Measure value 0.86, Matthews connection coefficient value 0.68. The Attribute Selected Classifier with Naïve Bayes Updateable of second order ensemble model gives highest performance which as ROC value 0.89 and PRC value 0.89, MAE value 0.15, RMSE value 0.40, RAE value 47.32%, RRSE value 90.11% and it takes time consumption as 0.09 seconds to build a model which is produced an optimal results based on their performance compare with other models. This Attribute Selected Classifiers with Naïve Bayes Updateable model is performing well compare with other models.

show abstract

Machine learning-based ship detection and tracking using satellite images for maritime surveillance

Cited by 38 publications

References 21 publications

Hash-Based Deep Learning Approach for Remote Sensing Satellite Imagery Detection

Hash-Based Deep Learning Approach for Remote Sensing Satellite Imagery Detection

A Novel Model for Network Intrusion Detection by Using Statistical Learning Approach

A Second Order Inductive Deduction Approach for Identification of Room Occupancy by Wireless Sensors

Contact Info

Product

Resources

About