Teledyne DALSA is an international leader in high performance digital imaging and semiconductors with approximately 1000 employees world-wide. Established in 1980 and acquired by Teledyne Technologies in 2011, Teledyne DALSA designs, develops, manufactures, and markets digital imaging products and solutions, in addition to providing semiconductor products and services

Teledyne Dalsa has now introduced an amazing Software platform Shape Extraction feature of Sherlock 7. This software offers a solution for extraction and inspection on parts or assemblies based on 3 dimensional shapes. These features include reading the embossed and engraved characters or marking on the metals.

This software combines multi-directional lighting with advanced software algorithm to eliminate surface background such as noises and colors and produces an image focused on the features which is relevant to the inspections. This output image can then be inspected using Sherlock vision tool. Sherlock SE is the most flexible and economical software available today.

Sherlock SE extracts the 3D surface structure of parts using process known as shape from shading. The same process is used for removing glares from high reflective parts.

The components used in Sherlock SE solution are PC or Vision platform running Sherlock 7.2.7.7 or higher. – 4-segment light setup in a circular, rectangular or square configuration. – Multi-segment light controller with Sherlock plugin license. – Genie Nano camera. – Associated cables and mechanical fixtures. CCS America has developed the following micro website which includes videos showing how to setup a system using Sherlock:

Application Demo:

Reading Braille on a medicine package

braille on a medicine package

  
Source image                                                                                      Computed Image

Software Analysis


Computed Image


Object Segmentation


Imperfects dots might difficult to train

Imperfect dots are converted in to perfect dots

Perfect dot character can easily read with OCR

Leave a Reply

Your email address will not be published. Required fields are marked *

Post comment