[dsm_typing_effect before_typing_effect=”Defect Detection ” typing_effect=”using Computer Vision” typing_cursor_color=”RGBA(255,255,255,0)” _builder_version=”4.22.1″ _module_preset=”default” before_font=”|600|||||||” before_text_color=”#0a2471″ before_font_size=”56px” typing_font=”|600|||||||” typing_text_color=”#0a2471″ typing_font_size=”56px” typing_line_height=”1.2em” after_font_size=”57px” hover_enabled=”0″ global_colors_info=”{}” sticky_enabled=”0″ theme_builder_area=”post_content”][/dsm_typing_effect]
US city with the largest service sewer pipe infrastructure.

Problem

Identify the defects in pipes. This city wanted a technology partner to annotate the sewer images at scale so that they could train a model to accurately detect the issues with the sewer pipes in the city.

The Intellekt Approach

Video and Image Annotations – The inspected images of anomalies in the sewers are annotated using bounding boxes and polygonal annotations.

Defect Categorization – These annotations are further classified under various domains pertaining to the faults in the sewer pipes like cracks, fractures, roots etc.

Result From The Customer

These annotations help the client build machine learning models which help the teams repair the anomalies of the underground sewer pipes.

Services Provided

Bounding Box Annotation

Image Annotation

Data Labelling

Model Training

Annotations to date

%

Accuracy Level