ImageJ Software
ImageJ is an open-source software, known for its wide user community, which has developed various plugins to easily execute different image processing algorithms (Introduction to ImageJ Image Processing Software). However, the software’s built-in set of basic tools already includes many essential image editing and analysis operations for users by default.
Shape Description of Objects
Shape is a key feature of each object. In various sectors of the food industry, particularly in the design and development of food machinery, there is a wide range of shapes related to raw materials and products. Some objects have a defined shape. For example, a piece of pasta could be described as cylindrical or a rectangle in two-dimensional terms.
However, for products like seeds, nuts, fruits, or pieces of meat and fish, no specific shape can be defined. In food biophysics, shape description is a crucial topic. Shape description might be done using a sphericity coefficient or by determining the major and minor diameters or fitting an elliptical shape to the object. This information is useful in food production, processing, and equipment design.
Imaging
Imaging can be performed on a single object or on a set of objects together. For instance, if the product, such as an apple, is large, each apple might be imaged separately in the study. In cases where the sample is smaller, like rice grains or almonds, a mass of the product may be imaged together. To achieve better sample separation, it is advised to use a background with a completely contrasting color or a white background (Application of Image Processing in the Food Industry).
Setting Scale in ImageJ
After obtaining and opening the images in the software, the first step is to set the scale. A ruler should be included in one of the images. Then, a specific portion of the ruler is selected, and from the Image menu, the “Scale” option is used to set the scale globally. This ensures that all measurements are based on the set scale.
Object Segmentation in ImageJ
It is important to note that the methods outlined are applicable to images with a contrasting and appropriate background. In other types of images, further preprocessing may be necessary. In the target image, first, the image type should be set to 8-bit from the “Type” option under the Image menu. This type configures the image with 256 different shades. Next, under the “Adjust” submenu of the “Image” menu, the “Threshold” option should be selected. Depending on the image, the shading range should be adjusted so that objects are separated from the background. Finally, the “Apply” option should be selected to confirm the change.
Converting Image to Binary
The next step is to convert the segmented image into binary form by selecting “Make Binary” under the “Binary” submenu in the “Process” menu. Other options like “Close” and “Fill Holes” in the same submenu can enhance the image. It is worth mentioning that options like “Dilute,” “Erode,” and “Watershed” can be effective in better object segmentation, depending on the image conditions.
Particle Analysis
In this step, first, under the “Analyze” menu, choose “Set Measurements” and activate the options for “Shape Descriptor,” “Feret’s Diameter,” and “Display Label.” Next, select “Analyze Particles” from the same menu and adjust the necessary settings. One important setting is determining the minimum size to exclude unwanted particles that were not separated during segmentation. The options “Show Outlines,” “Display Results,” and “Summary Results” can also be helpful for summarizing results. Once the window is confirmed, the particles will be identified, and shape parameters will be determined.
Interpreting Results
In the results window, each column is labeled accordingly. The “Circ” parameter represents the roundness, “AR” is the ratio of the major to minor diameter, and “Round” refers to the smoothness of the edges. Additionally, the “Feret” parameters are provided. The “Feret Diameter” corresponds to the maximum distance between two points on the object’s edge in a specified direction, while “Min Feret” represents the minimum distance between two points on the edge in a specific direction.
