Sawmill Industry

Project: Imagery defect tracking (computer vision programming)

Client : Barette Chapais Ltee

Technologies: C#, Halcon, .Net, scientific numerical programming

Industry : (sawmill facility) (lumber …)

Project Imagery processing (sawmills industry)
Wrote image processing software using third party library (Halcon) and .Net framework for a sawmill company. Programs written feature an algorithm and image concepts, defined as a sequence of instructions that operates on a set of input image, transforming that information into an output image that is of interest to the user. Developed a useful GUI for rapidly debugging code.
Technology used: Halcon image processing library, algorithm development, scientific programming,

Introduction

Hardwood sawmills major problem is optimizing the cuts for each board. Typical process consist logs are first cut into boards that retain residual bark, called wane, and other undesirable features of the log.

Tasks human based but it is something tha could be automated by using technolgy product. Automatic …optimizing the value (cuts) of each board about the placement based on their knowledge of lumber grades and current lumber prices. Ideally, edge and trim cuts will be selected so that the sale value of each board is maximized.

In practice, however, optimizing the value of each board in this manner is a complex decision that is difficult even for experienced operators.

The decision is based on the position of wane and defects, as well as current market prices for wood.

system that uses a structured-light system to obtain profile (thickness) images of unplaned boards, in addition to gray-scale images for defect detection

set of profile iamages and apply some filter (image processing) to extract wane (edge defect).

Gray scale image technology … scanning system captures profile images (or, equivalently, range images) in which each pixel represents a measurement of board thickness.

Our approach to the detection of wane is to search for discontinuities in surface characteristics of the wood

Wane detection and analysis is one part of an automated scanning system for optimizing edging and trimming of hardwood lumber in sawmills. This capability, together with automatic defect detection, will form the nucleus of computer-driven edging and trimming.

use with a structured-light profiling system … the scanning system captures profile images (or, equivalently, range images) in which each pixel represents a measurement of board thickness.

As an example, a portion of a profile image is given in Figure 1

Project Imagery processing (sawmills industry)
Wrote image processing software using third party library (Halcon) and .Net framework for a sawmill company. Programs written feature an algorithm and image concepts, defined as a sequence of instructions that operates on a set of input image, transforming that information into an output image that is of interest to the user. Developed a useful GUI for rapidly debugging code.

Project

Because of hardwood breakdown of hardwood logs into lumber produces boards with rough surface. These boards contain wane (missing wood du to the curved log exterior) that is removed by edge and trim cuts prior to sale. Because hardwood lumber value is determined using a combination of board size and quality, knowledge of wane position and defects is essential for selecting cuts that maximize profit.
HDevelop prototype (Halcon Tool)

Designed and implemented an object-oriented library to detect wane on rough hardwood lumber.

This library featured an open architecture, improving drastically on traditional technique. This library is used in the sawmill facility in the optimization process (cut optimization).

Wane definition?? missing wood on the edge of the piece.

Written an stand-alone application that implement an algorithm (image processing based) from scan points. These scan are converted into an image that can managed (handled) by image processing profile (find …)

System is based profile (thickness) images of unplanned board, in addition to gray-scale images for defect detection.
Detecting wane boundaries through the analysis of these profile images.

Defect Tracking Algorithm

defect tracking based on mathematics algorithm () Implemented an image processing algorithm for defect tracking.

operates on images, transforming in set of an output images that are of interest for the user. to detect wane (wood defect) on a rough hardwood lumber.

C# and .Net framework …

Validation Wane Algo

  1. Read shape file .shp (shape)
  2. Shape correction (detorsion board)
  3. Wane detection
  4. Display and write des valeurs de flache (text file)

Numerical Library fit () in

C (mixed language) library of math functions for fit () … to be completed

Programming Environment

  1. WaneAlgo library
  2. Numerical Library
  3. … to be completed
  4. GUI for debugging and validate

Wane Simulator to visualize wane (help to bvalidate and debug the algorithm). Steps consists of reading a shape file which contains …

Gray scale image representation of piece of wood ()

Developed a useful GUI for quick code debugging ()