Upper gastrointestinal endoscopies are primarily performed to observe the pathologies of the esophagus, stomach, and duodenum. However, when an endoscope is pushed into the esophagus or stomach by the physician, the organs behave similar to a balloon being gradually inflated. Consequently, their shapes and depth-of-field of images change continually, preventing thorough examination of the inflammation or anabrosis position, which delays the curing period. In this study, a 2.9-mm image-capturing module and a convoluted mechanism was incorporated into the tube like a standard 10- mm upper gastrointestinal endoscope. The scale-invariant feature transform (SIFT) algorithm was adopted to implement disease feature extraction on a koala doll. Following feature extraction, the smoothly varying affine stitching (SVAS) method was employed to resolve stitching distortion problems. Subsequently, the real-time splice software developed in this study was embedded in an upper gastrointestinal endoscope to obtain a panoramic view of stomach inflammation in the captured images. The results showed that the 2.9-mm image-capturing module can provide approximately 50 verified images in one spin cycle, a viewing angle of 120° can be attained, and less than 10% distortion can be achieved in each image. Therefore, these methods can solve the problems encountered when using a standard 10-mm upper gastrointestinal endoscope with a single camera, such as image distortion, and partial inflammation displays. The results also showed that the SIFT algorithm provides the highest correct matching rate, and the SVAS method can be employed to resolve the parallax problems caused by stitching together images of different flat surfaces.
The absorbing filter  is an optical element employed for isolating regions of a spectrum. In general, the thicker the absorbing filter material, the more wavelengths it will absorb. However, most optical filter products ignore light diffusion and are made with a constant thickness. While the non-collimated beams pass through the filter, the optical paths vary with incident angles. Thus, the absorption difference happens and leads to the poor uniformity of transmission spectrum. In our work, a filter lens was developed to achieve the similar function of interference filter and ND filter with better spectrum uniformity. It is mounted onto a designed macro lens and supplies it with a good spectrum aberration correction. The shape of the filter lens is designed to eliminate the optical path differences between the light beams in the medium. The macro lens is made of neutral glass and shaped into symmetrical biconvex for achieving macro imaging. The spectrum characteristic of the filter lens depends on the material of the absorbing filter. In the experiment, the filter lens was prepared. The experimental results show that the spectrum uniformity of the filter lens is better than that of the normal filter.
The skin illuminated of two lights at different wavelength can be applied to detect the oxygen saturation of human blood. Due to the absorption coefficient of oxy- (HbO2) and deoxy- (Hb) hemoglobin are different at the wavelength 660 nm and 890 nm, the transmitted and reflected light within the skin can be used to compute the oxygen saturation image of skin. However, the intensities of skin images illuminated by a 20 mW NIR-LED are too low to determine the position of blood vessel when acquired by the color CCD camera. In order to improve the disadvantages, a mono camera was used and the irradiated distance and angle between LED light and test hand were adjusted to acquire the higher resolution and contrast blood vessel images for the oxygen saturation calculation. In the experiment, we developed the suitable angle to irradiate NIR light is at 75 degrees because the reflected and scattered effect could be generated significantly from both vertical and horizontal direction. In addition, the best contrast vessel images can be obtained when the shutter time is set at 44.030 ms and the irradiated distance was at the range 140-160 mm due to the intensity ratio between tissue and vessel region is the highest and the intensities of image would not be saturated or become too low when these two parameters were adjusted slightly. In future, the proposed parameters and results can be applied to the oxygen saturation measurement in the clinical diagnosis.
Fluorescence objects can be excited by ultraviolet (UV) light and emit a specific light of longer wavelength in
biomedical experiments. However, UV light causes a deviation in the blue violet color of fluorescent images. Therefore,
this study presents a color deviation adjustment method to recover the color of fluorescent image to the hue observed
under normal white light, while retaining the UV light-excited fluorescent area in the reconstructed image. Based on the
Gray World Method, we proposed a non-linear logarithm method (NLLM) to restore the color deviation of fluorescent
images by using a yellow filter attached to the front of a digital camera lens in the experiment. Subsequently, the
luminance datum of objects can be divided into the red, green, and blue (R/G/B) components which can determine the
appropriate intensity of chromatic colors. In general, the datum of fluorescent images transformed into the CIE 1931
color space can be used to evaluate the quality of reconstructed images by the distribution of x-y coordinates. From the
experiment, the proposed method NLLM can recover more than 90% color deviation and the reconstructed images can
approach to the real color of fluorescent object illuminated by white light.
The fluorescent reaction is that an organism or dye, excited by UV light (200-405 nm), emits a specific frequency of
light; the light is usually a visible or near infrared light (405-900 nm). During the UV light irradiation, the photosensitive
agent will be induced to start the photochemical reaction. In addition, the fluorescence image can be used for
fluorescence diagnosis and then photodynamic therapy can be given to dental diseases and skin cancer, which has
become a useful tool to provide scientific evidence in many biomedical researches. However, most of the methods on
acquiring fluorescence biology traces are still stay in primitive stage, catching by naked eyes and researcher's subjective
judgment. This article presents a portable camera to obtain the fluorescence image and to make up a deficit from
observer competence and subjective judgment. Furthermore, the portable camera offers the 375nm UV-LED exciting
light source for user to record fluorescence image and makes the recorded image become persuasive scientific evidence.
In addition, when the raising the rate between signal and noise, the signal processing module will not only amplify the
fluorescence signal up to 70 %, but also decrease the noise significantly from environmental light on bill and nude mouse