Apply IMAGING SYSTEM and mechanical devices for food sorting

Imaging technologies: Near infrared spectroscopy
Food control parameters: Structure of molecules, instrumental texture, sensory tenderness, pastiness, crusting, juiciness, discrimination between fresh and frozen-thawed products, water holding capacity

In contrast to conventional methods for the determination of meat chemical composition and quality,
near infrared spectroscopy (NIRS) enables rapid, simple and simultaneous assessment of numerous meat properties.

source: wiki
Technology using near-IR absorption spectrum of dichloromethane showing complicated overlapping overtones of mid IR absorption features.

Optical microscopy
Food control parameters: Fat organization, collagen organization, myofiber typing, discrimination between fresh and frozen-thawed products, myofiber spacing, Z line degradation, sarcomere length, endomysium structure, myofiber diameter, myofiber density, myofiber organization, PSE, specific proteins detection, collagen typing, myofilaments organization

Image Visualization Platform
Nauru Technology is a platform and drivers for digital image data management and image processing . Using Nauru you can now visualize, organize, browse, process, analyse your large image repositories (files, databases) from microscope, radiology, astronomy, etc in quick, simple steps. It can read many image formats including TIFF, GIF, JPEG, BMP, DICOM, FITS and ‘raw’. It can handle unlimited amount of images in interactive user interface. IM can be installed on every computer and can analyse image storage in real time, providing sophisticated reports.

Main features:

-        Used as local database system

Nauru can read unlimited amount of images from file storage, manage & filter them, search, interactive visualize and save search setting as profile in project. IM can export desired image collection into new location.

-        Used as image storage miner

Nauru  can control and monitor in amount of images and folders created on hard drive or network share drive. Nauru can provide full report as a table about current storage conditions: amount of files, folders, storage size risk. IM can export as table an view of entire image storage.  

Nauru  can inform users about different type of image data, size, duplicates, creation date, creation by, last access.  Nauru provide time based report: amount of images by year, month day, time statistics

-        Used as visual analytics tool

Nauru  can visual analyse data by providing interactive plots (scatter, bar, line, pie) and filter mechanism connected to report tables

-        Used for interactive image data labelling and teaching module for machine learning: Nauru can be used to interactive labelling of images by generating in data table a new column with new class. Interactive labeling is used very often for manual data annotation, extending image metadata or for preparation of the database for classification using machine learning algorithms

-        Used for automated image processing

Nauru can perform image processing using algorithms implemented in Nauru based on ImageJ or by request implement any additional image processing algorithm from any image processing technology or programming language

-         Used for classification

Images can be automated classified using machine learning algorithms

-          Nauru is available as KNIME node

Possible usage scenario of Nauru:

·         Find similar images on image storage
·         Find identical images on image storage
·         Create a full table of images existing on storage -> export to excel or text file for further analysis using external tools
·         Create a full table of images existing on storage and expand the table by image metadata and parameter -> export to excel or text file for further analysis using external tools
·         Analyse images by content based using image metadata
·         Check statistics for different type of images: amount of images for each type
·         Check which folder on my storage contains largest collection of images
·         Check which folder contain largest images
·         Find using scatter plot a group of images of interest: pattern
·         Find using scatter plot largest images on storage
·         Search for images using multiparamteric filters
·         Control over time production of images on storage
·         Batch image processing (change size, color, threshold, type: 8; 16; 32 bits, format, resolution, …. ) -> export processed images to new folder
·         Label images by creating new column in the table
·         Find images by selecting them in scatter plot -> navigate to folders, files
·         Save current imported images, current view, current filter settings, selected images as project and come back to same setting loading the project again.