could be lossy and lossless, for example, TIFF (tagged image file format) is
designed as a container to hold uncompressed image, lossy compressed image, or
lossless compressed image.
Image segmentation: Image segmentation partitions an image into multiple segments. The goal of this stage is to identify the object in a digital image. Sometimes digital images we deal with are very complicated; thus, the segmentation
could be a tough process in the digital image processing system. Basic
approaches and methods applied in image segmentation includes sharp (point,
line, edge) detection, thresholding methods such as global thresholding and
Otsu’s method, color-based segmentation such as K-means clustering, transform
methods such as morphological watersheds, region-based segmentation, and
texture filter.
Image representation: Image representation deals with the problem of how the
image is represented. Generally, the process of image representation consists of
two stages: choosing representation scheme and describing the region based on
the representation. After the processing, the image would be represented with a
proper description.
6.3 Hardware and Software
A typical image processing system consists of hardware and software. The hardware
could be a personal computer (PC), an image sensor, or a specialized image
processing hardware. The software could be a software package which is capable
of running on a specific platform to process specific task or a library providing
application programming interface (API) to facilitate image processing–related
development and implementation. This section covers a suite of common image
processing hardware/software as well as state-of-the-art technologies such as mobile
device–based image processing and cloud-based image processing.
6.3.1 Image Processing Hardware
The digital image processing technology is evolving with the development and
widespread of computer hardware technology. In the mid-1980s, digital image
data were analyzed in microcomputer-based image analysis systems comprising
linked components with a monitor for displaying user interface and another for
displaying images. With the widespread of personal computer (PC) since the 1990s,
more and more digital image processing systems are developed based on desktop
platform with graphic user interfaces (GUI). In the recent years, with the quickly
development of the Internet technologies and infrastructures, more platforms and
concepts such as cloud computing, mobile computing, and machine learning frameworks are being applied in image processing.
6 Image Processing Methods in Agricultural Observation Systems
85
designed as a container to hold uncompressed image, lossy compressed image, or
lossless compressed image.
Image segmentation: Image segmentation partitions an image into multiple segments. The goal of this stage is to identify the object in a digital image. Sometimes digital images we deal with are very complicated; thus, the segmentation
could be a tough process in the digital image processing system. Basic
approaches and methods applied in image segmentation includes sharp (point,
line, edge) detection, thresholding methods such as global thresholding and
Otsu’s method, color-based segmentation such as K-means clustering, transform
methods such as morphological watersheds, region-based segmentation, and
texture filter.
Image representation: Image representation deals with the problem of how the
image is represented. Generally, the process of image representation consists of
two stages: choosing representation scheme and describing the region based on
the representation. After the processing, the image would be represented with a
proper description.
6.3 Hardware and Software
A typical image processing system consists of hardware and software. The hardware
could be a personal computer (PC), an image sensor, or a specialized image
processing hardware. The software could be a software package which is capable
of running on a specific platform to process specific task or a library providing
application programming interface (API) to facilitate image processing–related
development and implementation. This section covers a suite of common image
processing hardware/software as well as state-of-the-art technologies such as mobile
device–based image processing and cloud-based image processing.
6.3.1 Image Processing Hardware
The digital image processing technology is evolving with the development and
widespread of computer hardware technology. In the mid-1980s, digital image
data were analyzed in microcomputer-based image analysis systems comprising
linked components with a monitor for displaying user interface and another for
displaying images. With the widespread of personal computer (PC) since the 1990s,
more and more digital image processing systems are developed based on desktop
platform with graphic user interfaces (GUI). In the recent years, with the quickly
development of the Internet technologies and infrastructures, more platforms and
concepts such as cloud computing, mobile computing, and machine learning frameworks are being applied in image processing.
6 Image Processing Methods in Agricultural Observation Systems
85