scholarly journals Analysis of the Repeatability of SIFT and SURF Descriptors Techniques for Underwater Image Preprocessing

2021 ◽  
pp. 24-31
Author(s):  
Shubhangi N. Ghate ◽  
Dr. Mangesh Nikose
Keyword(s):  
Prior Knowledge ◽  
State Of The Art ◽  
Rotation Invariant ◽  
Image Preprocessing ◽  
Retrieval Time ◽  
Retrieval Accuracy ◽  
Invariant Features ◽  
Underwater Image ◽  
Geometrical Constraints ◽  
Time Required

To improve the repeatability of SIFT and SURF descriptors, we conducted research to find two methods: first, a method for pre-processing underwater images that does not require prior knowledge of the scene, and second, a method for computing distances that is less expensive in terms of execution time for finding corresponding points. SIFTs (Scale and Rotation Invariant Features) are new features that have been developed. SIFTs (Scale and Rotation Invariant Features) are newly developed features that are based on geometrical constraints between pairs of nearby points around a key point. SIFT is contrasted with cutting-edge local features. SIFT outperforms the state-of-the-art in terms of retrieval time and retrieval accuracy. We have discussed the time required to extract key point features of SIFT and SURF Descriptor.

JIRL

2013 ◽  
Vol 4 (2) ◽  
pp. 1-12 ◽  
Author(s):  
David Edmundson ◽  
Gerald Schaefer
Keyword(s):  
Image Retrieval ◽  
Comparative Evaluation ◽  
State Of The Art ◽  
Object Oriented ◽  
Compressed Domain ◽  
Software Suite ◽  
Retrieval Time ◽  
Jpeg Images ◽  
Retrieval Accuracy ◽  
Retrieval Algorithms

Since there are few open image retrieval toolkits available, researchers in the field are often forced to re-implement existing algorithms in order to perform a comparative evaluation. None of the existing toolkits support retrieval of JPEG images directly in the compressed domain. The authors’ aim is therefore to facilitate the use of compressed domain image retrieval techniques as well as ease retrieval evaluation by fellow researchers. For this purpose, the authors present JIRL, an open source C++ software suite that allows content-based image retrieval in the JPEG compressed domain and provides tools for benchmarking retrieval accuracy and retrieval time. In total, twelve state-of-the-art JPEG retrieval algorithms are implemented, while for each method techniques for compressed domain feature extraction as well as feature comparison are provided in an object-oriented framework. An example image retrieval application is also provided to demonstrate how the library can be used. JIRL is made available to fellow researchers under the LGPL v.2.1 license.

scholarly journals Label Enhancement for Label Distribution Learning via Prior Knowledge

2020 ◽  
Author(s):  
Yongbiao Gao ◽  
Yu Zhang ◽  
Xin Geng
Keyword(s):  
Machine Learning ◽  
Reinforcement Learning ◽  
Emotion Recognition ◽  
Prior Knowledge ◽  
Decision Process ◽  
Age Estimation ◽  
State Of The Art ◽  
Learning Agent ◽  
Label Distribution Learning ◽  
Label Distribution

Label distribution learning (LDL) is a novel machine learning paradigm that gives a description degree of each label to an instance. However, most of training datasets only contain simple logical labels rather than label distributions due to the difficulty of obtaining the label distributions directly. We propose to use the prior knowledge to recover the label distributions. The process of recovering the label distributions from the logical labels is called label enhancement. In this paper, we formulate the label enhancement as a dynamic decision process. Thus, the label distribution is adjusted by a series of actions conducted by a reinforcement learning agent according to sequential state representations. The target state is defined by the prior knowledge. Experimental results show that the proposed approach outperforms the state-of-the-art methods in both age estimation and image emotion recognition.

Contained Power Morcellation Versus Transvaginal Extraction for Retrieval of Laparoscopically Removed Myomas: A Comparison of Perioperative Outcomes

2018 ◽  
Vol 26 (1) ◽  
pp. 72-76 ◽  
Author(s):  
Aysen Boza ◽  
Selim Misirlioglu ◽  
Cagatay Taskiran ◽  
Bulent Urman
Keyword(s):  
Hospital Stay ◽  
Hospital Costs ◽  
Final Analysis ◽  
Perioperative Outcomes ◽  
Retrieval Time ◽  
Operative Outcomes ◽  
Specimen Retrieval ◽  
Estimated Blood Loss ◽  
Power Morcellation ◽  
Time Required

Objective. To evaluate clinical and operative outcomes of transvaginal extraction (TVE) and contained power morcellation (CPM) for myoma retrieval after laparoscopic myomectomy. Materials and Methods. Prospective data from 35 consecutive cases using CPM were compared with retrospective data of all cases using TVE from December 2014 to January 2017. Patients were matched 1:1 based on myoma diameter. A total of 62 women were included in the final analysis. Specimen retrieval was performed using the TVE or CPM within an insufflated isolation bag. Results. Age, body mass index, mode of prior obstetric delivery, history of previous abdominal surgery, indication for myomectomy, and the myoma(s) characteristics were similar between groups. Retrieval time was significantly shorter in the TVE group compared with the CPM group: 10 minutes (3-15 minutes) versus 17 minutes (14-42 minutes); P < .001. Time required for placement of the instruments was 9.7 minutes for the isolation bag and 0.5 minutes for the vaginal extractor. Additional analgesic administration for pain relief was necessary in 13 patients (42%) in the TVE group and 23 patients (72%) in the CPM group ( P = .01). Total cost of the hospital stay was significantly higher in the CPM group compared with the TVE group ( P < .001). Estimated blood loss and duration of hospital stay were similar between groups. Conclusion. Both CPM and TVE can be used for safe retrieval of large myomas that are removed laparoscopically. Compared with CPM, TVE was associated with a shorter retrieval time, less postoperative pain, and less hospital costs.

scholarly journals Frame-Based Facial Expression Recognition Using Geometrical Features

2014 ◽  
Vol 2014 ◽  
pp. 1-13 ◽  
Author(s):  
Anwar Saeed ◽  
Ayoub Al-Hamadi ◽  
Robert Niese ◽  
Moftah Elzobi
Keyword(s):  
Facial Expression ◽  
Prior Knowledge ◽  
State Of The Art ◽  
Recognition Rate ◽  
Human Interaction ◽  
Expression Recognition ◽  
Acoustic Data ◽  
Geometrical Features ◽  
Benchmark Datasets ◽  
Neutral Expression

To improve the human-computer interaction (HCI) to be as good as human-human interaction, building an efficient approach for human emotion recognition is required. These emotions could be fused from several modalities such as facial expression, hand gesture, acoustic data, and biophysiological data. In this paper, we address the frame-based perception of the universal human facial expressions (happiness, surprise, anger, disgust, fear, and sadness), with the help of several geometrical features. Unlike many other geometry-based approaches, the frame-based method does not rely on prior knowledge of a person-specific neutral expression; this knowledge is gained through human intervention and not available in real scenarios. Additionally, we provide a method to investigate the performance of the geometry-based approaches under various facial point localization errors. From an evaluation on two public benchmark datasets, we have found that using eight facial points, we can achieve the state-of-the-art recognition rate. However, this state-of-the-art geometry-based approach exploits features derived from 68 facial points and requires prior knowledge of the person-specific neutral expression. The expression recognition rate using geometrical features is adversely affected by the errors in the facial point localization, especially for the expressions with subtle facial deformations.

Sign in / Sign up

Export Citation Format

Share Document