Deep learning has achieved great accuracy in large scale image classification and scene recognition tasks, especially after the Convolutional Neural Network (CNN) model was introduced. Although a CNN often demonstrates very good classification results, it is usually unclear how or why a classification result is achieved. The objective of this thesis is to explore several existing visualization approaches which offer intuitive visual results. The thesis focuses on three visualization approaches: (1) image masking which highlights the region of image with high influence on the classification, (2) Taylor decomposition back-propagation which generates a per pixel heat map that describes each pixel's effect on the classification, and (3) Inception which generates a natural looking image based on the features maximizing the classification score. We explore two challenging visualization tasks, (1) visualizing a model classifying images based on the time when they are taken, and (2) visualizing a model of predicting plant phenotypes (specifically wheat heading percentage). The thesis demonstrates how these visualization approaches work for both the classification model and regression model, and evaluates the results on real-world imagery.
","abstract_html":"<p>Deep learning has achieved great accuracy in large scale image classification and scene recognition tasks, especially after the Convolutional Neural Network (CNN) model was introduced. Although a CNN often demonstrates very good classification results, it is usually unclear how or why a classification result is achieved. The objective of this thesis is to explore several existing visualization approaches which offer intuitive visual results. The thesis focuses on three visualization approaches: (1) image masking which highlights the region of image with high influence on the classification, (2) Taylor decomposition back-propagation which generates a per pixel heat map that describes each pixel's effect on the classification, and (3) Inception which generates a natural looking image based on the features maximizing the classification score. We explore two challenging visualization tasks, (1) visualizing a model classifying images based on the time when they are taken, and (2) visualizing a model of predicting plant phenotypes (specifically wheat heading percentage). The thesis demonstrates how these visualization approaches work for both the classification model and regression model, and evaluates the results on real-world imagery.</p>","abstract_has_math":false,"creators":["Li, Dingwen"],"institution":null,"degree_name":"Master of Science (MS)","degree_level":"Thesis","degree_discipline":"Computer Science & Engineering","degree_department":null,"school":null,"contributors":["Robert Pless","Yasu Furukawa Tao Ju"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2016,"date_issued":"2016-05-15T07:00:00Z","date_published":"2016-05-15T07:00:00Z","updated_at":"2026-07-24T06:13:14Z","subjects":["Convolutional Neural Network","Deep Learning","Machine Learning","Visualization","Image Classification","Engineering","Other Computer Sciences"],"languages":["English (en)"],"rights":["I have not registered my thesis with the U.S. Copyright Office, and do not intend to."],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["https://openscholarship.wustl.edu/eng_etds/150"],"render_values":[{"text":"https://openscholarship.wustl.edu/eng_etds/150","href":"https://openscholarship.wustl.edu/eng_etds/150","code":true}]}]},"links":{"outbound_url":"https://doi.org/10.7936/K7BK19NT","outbound_label":"DOI","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Robert Pless","Yasu Furukawa Tao Ju"]},{"key":"dc:creator","label":"Author","values":["Li, Dingwen"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.available","label":"Dc Date Available","values":["2016-05-05T07:00:00Z"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Computer Science & Engineering","McKelvey School of Engineering"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Thesis"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master of Science (MS)"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Convolutional Neural Network","Deep Learning","Machine Learning","Visualization","Image Classification","Engineering","Other Computer Sciences"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["English (en)"]},{"key":"dc:rights","label":"Dc Rights","values":["I have not registered my thesis with the U.S. Copyright Office, and do not intend to."]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://doi.org/10.7936/K7BK19NT","https://openscholarship.wustl.edu/eng_etds/150"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Permanent URL: https://doi.org/10.7936/K7BK19NT"]},{"key":"dc:description.abstract","label":"Abstract","values":["Deep learning has achieved great accuracy in large scale image classification and scene recognition tasks, especially after the Convolutional Neural Network (CNN) model was introduced. Although a CNN often demonstrates very good classification results, it is usually unclear how or why a classification result is achieved. The objective of this thesis is to explore several existing visualization approaches which offer intuitive visual results. The thesis focuses on three visualization approaches: (1) image masking which highlights the region of image with high influence on the classification, (2) Taylor decomposition back-propagation which generates a per pixel heat map that describes each pixel's effect on the classification, and (3) Inception which generates a natural looking image based on the features maximizing the classification score. We explore two challenging visualization tasks, (1) visualizing a model classifying images based on the time when they are taken, and (2) visualizing a model of predicting plant phenotypes (specifically wheat heading percentage). The thesis demonstrates how these visualization approaches work for both the classification model and regression model, and evaluates the results on real-world imagery.
"]},{"key":"dc:title","label":"Title","values":["Visualization of Deep Convolutional Neural Networks"]}]}],"canonical_facts":{"dc:contributor":["Robert Pless","Yasu Furukawa Tao Ju"],"dc:creator":["Li, Dingwen"],"dc:date.available":["2016-05-05T07:00:00Z"],"dc:description":["Permanent URL: https://doi.org/10.7936/K7BK19NT"],"dc:description.abstract":["Deep learning has achieved great accuracy in large scale image classification and scene recognition tasks, especially after the Convolutional Neural Network (CNN) model was introduced. Although a CNN often demonstrates very good classification results, it is usually unclear how or why a classification result is achieved. The objective of this thesis is to explore several existing visualization approaches which offer intuitive visual results. The thesis focuses on three visualization approaches: (1) image masking which highlights the region of image with high influence on the classification, (2) Taylor decomposition back-propagation which generates a per pixel heat map that describes each pixel's effect on the classification, and (3) Inception which generates a natural looking image based on the features maximizing the classification score. We explore two challenging visualization tasks, (1) visualizing a model classifying images based on the time when they are taken, and (2) visualizing a model of predicting plant phenotypes (specifically wheat heading percentage). The thesis demonstrates how these visualization approaches work for both the classification model and regression model, and evaluates the results on real-world imagery.
"],"dc:identifier":["https://doi.org/10.7936/K7BK19NT","https://openscholarship.wustl.edu/eng_etds/150"],"dc:language":["English (en)"],"dc:rights":["I have not registered my thesis with the U.S. Copyright Office, and do not intend to."],"dc:subject":["Convolutional Neural Network","Deep Learning","Machine Learning","Visualization","Image Classification","Engineering","Other Computer Sciences"],"dc:title":["Visualization of Deep Convolutional Neural Networks"],"thesis:degree_discipline":["Computer Science & Engineering","McKelvey School of Engineering"],"thesis:degree_level":["Thesis"],"thesis:degree_name":["Master of Science (MS)"]},"updated_at":"2026-07-24T06:13:14Z"}