Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Dissociable somatotopic representations of Chinese action verbs in the motor and premotor cortex.

Scientific reports·2013

Same author

High quantum-yield CdSexS1-x/ZnS core/shell quantum dots for warm white light-emitting diodes with good color rendering.

Nanotechnology·2013

Same author

A novel microfluidic mixer based on dual-hydrodynamic focusing for interrogating the kinetics of DNA-protein interaction.

The Analyst·2013

Same author

Correlation of reorientational jumps of water molecules in bulk water.

Physical review. E, Statistical, nonlinear, and soft matter physics·2013

Same author

Palladium-catalyzed direct arylation of C-H bond to construct quaternary carbon centers: the synthesis of diarylfluorene.

Organic letters·2013

Same author

Strength and fracture behavior of graphene grain boundaries: effects of temperature, inflection, and symmetry from molecular dynamics.

Physical chemistry chemical physics : PCCP·2013

Search research articles

Related Experiment Video

Updated: Dec 22, 2025

Deep Neural Networks for Image-Based Dietary Assessment

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

[Deep learning network-based recognition and localization of diatom images against complex background].

Jiehang Deng^1,2, Dongdong He¹, Jiahong Zhuo¹

¹School of Computer Science, Guangdong University of Technology, Guangzhou 510006, China.

Nan Fang Yi Ke Da Xue Xue Bao = Journal of Southern Medical University

|May 8, 2020

Summary

A new deep learning method accurately recognizes and locates diatoms in complex autopsy backgrounds. This advanced technique significantly outperforms traditional methods, improving forensic analysis.

Keywords:

complex background deep learning diatom machine learning object detection

More Related Videos

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation

Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation

Published on: February 9, 2024

Related Experiment Videos

Last Updated: Dec 22, 2025

Deep Neural Networks for Image-Based Dietary Assessment

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation

Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation

Published on: February 9, 2024

Area of Science:

Forensic Science
Computational Biology
Image Analysis

Background:

Diatom analysis is crucial in forensic science for determining drowning.
Complex backgrounds in autopsy samples hinder accurate diatom recognition and localization.
Existing methods struggle with identifying diatoms amidst challenging visual noise.

Purpose of the Study:

To develop a deep learning network for robust diatom recognition and localization in complex autopsy backgrounds.
To improve the accuracy and reliability of diatom analysis in forensic investigations.

Main Methods:

A two-module system combining ZFNet for feature extraction and Regional Proposal Network (RPN) for initial localization.
Fast R-CNN was employed for refining diatom position and classification.
A custom database with varying background complexities was used for testing.

Main Results:

Conventional methods achieved only ~60% recognition with partial background interference and failed with complex backgrounds.
The proposed deep learning method achieved an average recognition rate of 85% against complex backgrounds.
The system successfully recognized and located diatom targets even in challenging environments.

Conclusions:

The proposed deep learning network effectively addresses the challenge of diatom recognition in complex autopsy backgrounds.
This method offers a significant advancement for forensic diatom analysis.
The system demonstrates high potential for practical application in forensic casework.