The Effectiveness of Using a Pretrained Deep Learning Neural Networks for Object Classification in Underwater Video

Szymak, Piotr; Piskur, Paweł; Naus, Krzysztof

doi:10.3390/rs12183020

Cited by 26 publications

(15 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Other applications of AI are related to swarm intelligence [ 121 , 122 ], neural simulation [ 123 , 124 ], and evolutionary computing [ 125 ] for controlling fleets of vehicles. Similarly, neural networks are often used to detect and localize underwater objects [ 126 , 127 ], fins sensors can inform the design of control systems of fin-driven robots [ 128 ], while bio-inspired algorithms are applied for localizing odour or chemical sources by underwater robots [ 129 ].…”

Section: Discussionmentioning

confidence: 99%

Research Trends and Future Perspectives in Marine Biomimicking Robotics

Aguzzi

Costa

Funari

et al. 2021

Sensors

View full text Add to dashboard Cite

Mechatronic and soft robotics are taking inspiration from the animal kingdom to create new high-performance robots. Here, we focused on marine biomimetic research and used innovative bibliographic statistics tools, to highlight established and emerging knowledge domains. A total of 6980 scientific publications retrieved from the Scopus database (1950–2020), evidencing a sharp research increase in 2003–2004. Clustering analysis of countries collaborations showed two major Asian-North America and European clusters. Three significant areas appeared: (i) energy provision, whose advancement mainly relies on microbial fuel cells, (ii) biomaterials for not yet fully operational soft-robotic solutions; and finally (iii), design and control, chiefly oriented to locomotor designs. In this scenario, marine biomimicking robotics still lacks solutions for the long-lasting energy provision, which presently hinders operation autonomy. In the research environment, identifying natural processes by which living organisms obtain energy is thus urgent to sustain energy-demanding tasks while, at the same time, the natural designs must increasingly inform to optimize energy consumption.

show abstract

Section: Discussionmentioning

confidence: 99%

Research Trends and Future Perspectives in Marine Biomimicking Robotics

Aguzzi

Costa

Funari

et al. 2021

Sensors

View full text Add to dashboard Cite

show abstract

“…The first approach uses a pretrained CNN as a feature extractor by removing the last fully connected layer and using a new dataset to train a new task classifier [21]. The second approach is by fine-tuning a pretrained CNN, training a new classification layer, and retraining several pre-trained CNN layers [22]. In this study, the pre-trained MobileNet V2 and NASNetMobile are finetuned to classify the banana ripeness.…”

Section: Training Cnnmentioning

confidence: 99%

Banana Ripeness Classification Based on Deep Learning using Convolutional Neural Network

Saragih

Emanuel

2021

2021 3rd East Indonesia Conference on Computer and Information Technology (EIConCIT)

View full text Add to dashboard Cite

Fruit ripeness is an important thing in agriculture because it determines the fruit's quality. Determining the ripeness of the fruit that was done manually poses several weaknesses, such as takes a relatively long time, requires a lot of labor, and can cause inconsistencies. The agricultural sector is one of the important sectors of the economy in Indonesia. However, sometimes the process of determining fruit ripeness is still done by using the manual method. The development of computer vision and machine learning technologies can be used to classify fruit ripeness automatically. This study applies the Convolutional Neural Network to classify the ripeness of the banana. The banana's ripeness is divided into four classes: unripe/green, yellowish-green, mid-ripen, and overripe. Two pre-trained models are used, which are MobileNet V2 and NASNetMobile. The experiment was conducted using Google Colab and several libraries such as OpenCV, Tensorflow, and scikit-learn. The result shows that MobileNet V2 achieves higher accuracy and faster execution time than the NASNetMobile. The highest accuracy achieved is 96.18%.

show abstract

“…To process the obtained spectra, ANNs were used. The earlier results of the research with the ANNs are included in [19,20].…”

Section: Introductionmentioning

confidence: 99%

Monitoring Time-Non-Stable Surfaces Using Mobile NIR DLP Spectroscopy

et al. 2022

Self Cite

View full text Add to dashboard Cite

In recent years, Near Infrared (NIR) spectroscopy has increased in popularity and usage for different purposes, including the detection of particular substances, evaluation of food quality, etc. Usually, mobile handheld NIR spectroscopy devices are used on the surfaces of different materials, very often organic ones. The features of these materials change as they age, leading to changes in their spectra. The ageing process often occurs only slowly, i.e., corresponding reflection spectra can be analyzed each hour or at an even longer interval. This paper undertakes the problem of analyzing surfaces of non-stable, rapidly changing materials such as waxes or adhesive materials. To obtain their characteristic spectra, NIR spectroscopy using a Digital Light Projection (DLP) spectrometer was used. Based on earlier experiences and the current state of the art, Artificial Neural Networks (ANNs) were used to process spectral sequences to proceed with an enormous value of spectra gathered during measurements.

show abstract

The Effectiveness of Using a Pretrained Deep Learning Neural Networks for Object Classification in Underwater Video

Cited by 26 publications

References 32 publications

Research Trends and Future Perspectives in Marine Biomimicking Robotics

Research Trends and Future Perspectives in Marine Biomimicking Robotics

Banana Ripeness Classification Based on Deep Learning using Convolutional Neural Network

Monitoring Time-Non-Stable Surfaces Using Mobile NIR DLP Spectroscopy

Contact Info

Product

Resources

About