Fluorescence light for contactless detection of living salmon lice on salmon skin
Kari Anne Hestnes Bakke*, Trine Kirkhus and Jon Tschudi
Sintef Digital, Smart sensors and micro systems, Forskningsveien 1, 0373 Oslo, Norway
* Corresponding author: kari.A.Bakke@sintef.no
Abstract
This work presents a promising method for automatic, contactless detection and counting of salmon lice that are infected on salmon in an aquaculture farm. The method uses fluorescence of chitin in the visual part of the spectrum to improve the contrast between fish skin and salmon lice and shows that fluorescence is even strong enough to give a real-time picture of the digestive and reproductive system in living lice without use of Dyes. The wavelengths used are compatible with an underwater measurement system.
Keywords: fluorescence / imaging / underwater / salmon louse
© De Autor (s), published by EDP Sciences, 2023
1 Introduction
One of the biggest threats for the salmon farm in Norway is a coppode, the Lepeophtheirus salmon lice Salmonis (L. Salmonis). It occurs naturally in seawater in the entire northern hemisphere. It lives and plans on salmon and trout in seawater, and lice are found in aquaculture locations as well as on trout and salmon in the fjords and along the coast, throughout the year. Lice pests in salmon can cause wounds in the skin of the fish and the welfare of the salmon will decrease if there are larger attacks. Over the years, the lice have developed resistance to various medical treatments and in some cases the only solution is slaughtering at a specific breeding location. In order to keep the well -being at a high level and to prevent the spread of salmon lice between locations, the salmon farmers have to perform liver racks on a representative group of salmon every week to ensure that the number of lice in the cage is less than less than is less than 0.5 fertile female lice per salmon. Numbers must be reported to the authorities. Systems for automatic census of lice are therefore highly sought after. However, performing living livering with the help of an underwater camera in the cage is demanding. The counting system must count the lice (2–6 mm) on the living salmon (60 cm) or visualize from 1 to 2 m distance through sea water, while swimming at a speed of approximately 1 fish length/sec. Other studies that focus on identifying and counting planktonic L. Salmonis stages in plankton samples in the laboratory. Thompson et al. Have investigated whether a specific combination of excitation and emission filters would result in L. Salmonis fluorescing, while the fluorescence of non-target white plankton is suppressed. The results are described by an excitation and emission matrix. However, the resulting matrix presented in [3] is limited to excitation under 580 Nm and emission under 600 Nm. Based on this, they present an excitation wave length of 380 Nm with emission detection in emission wave length 474 Nm or excitation at 450 Nm with detection at 516 Nm for identification of L. Salmonis larvae in laboratory image formation systems.
2 methods and experiments
In our studies, the fluorescence properties of salmon and lice were assessed to find a method to improve the image contrast between lice and skin, in a spectral area that can be applied under water. This will simplify the image analysis and the calculation needs and reduce the need for optical resolution, since the method is specific enough. We had access to newly slaughtered salmon with living lice that were delivered to seawater. Properties of lice, fish skin and seawater were investigated. The Vishid has large color variations and has stains in the colors white, gray and black. The lice themselves are semi-mirroring and partially transparent, and the color of the non-transparent part will vary between lice. The contrast between lice and fish skin in a normal camera image is therefore very limited. High resolution camera images of lice on fish skin are given in the air. Due to the low contrast, high -quality images are needed to identify the lice. It is very difficult to get images of sufficient quality (little noise and high resolution) in a real situation, and it would also be difficult to automate the process of identifying the lice. To improve the contrast, it is extra attractive to tackle the fluorescence properties, because the fluorescence depends on the difference in chemical properties of lice and salmon. The chemical substance of the co -epode of the lice differs from the oily salmon and its fish skin. Through our studies we discovered that red fluorescence emissions (> 600 Nm) could be obtained from lice using a 532 Nm, green, excitation wave length. This set of excitation/emission wave lengths also improves the contrast of salmon skin, since salmon skin fluorescence is not excited by the use of this wavelength [4]. An important point of attention for an underwater image formation system is also the absorption of light in the water column itself. Excitation at 380 Nm central wavelength with emissions above 410 Nm was tested in our lab. Our measurements confirmed that salmon lice have a fluorescence signal when they are excited in this wavelength oak. However, the wavelengths are not compatible with the other limitations in an image improving system that separates lice from salmon skin into an underwater image formation that works remotely. The contrast between lice and fish skin is not improved because the salmon skin also gives a fluorescence signal, and the absorption of UV light in seawater is unpredictable due to varying organic matter in the water column. The best transmission star in seawater is within the range of 500-700 Nm and the wavelengths that we propose are well suited for underwater use due to low absorption. Within the reach of 500 Nm-700 Nm, 80-95% of the light is passed on to the detector. This is in contrast with the wavelength oak under 425 Nm, where we have measured less than 20% through a water column of 1 m. Based on these findings, we have set up a fluorescence portrayal system. Semiconductor lasers at 532 Nm were used for excitation. A Long-Pass, glass filter with cut-on wavelength of 570 Nm was used for the Canon EOS M6 II camera to block the light of the laser, which resulted in the visible wavelengths above 570 Nm that reached the camera gip. A 28 mm macro lens was recorded for a high image resolution. Two lasers with different capacity were used for excitation. The diameter of the enlightened spot areas was approximately 4 mm and 12–14 mm, which gave a similar intensity for both lasers.
3 results
The lice are 3-4 mm in size and the field of view is approximately 10 mm x 15 mm. A white light source with an RGB camera was used for the proposed fluorescence imaging system. In the fluorescence image we can observe complex details of the salmon louse anatomy. We expected to observe the exoskeleton made of chitin, and were surprised by the detailed image of the louse anatomy. After observing these details, we learned from the literature that copepods have chitin not only in their exoskeleton, but also as part of their digestive and reproductive systems. These systems are clearly visible in the fluorescence images. The contrast between the lice and the salmon skin is high, as we could expect from our first fluorescence measurements. In our setup we used a rather long exposure time of 0.4–0.8 s and a high-resolution RGB camera. The quality of these images is much better than necessary for lice detection, both in terms of resolution and low image noise. A commercial system will have to compromise on image quality to obtain images of sufficient quality with a much shorter exposure time. In addition to the images obtained with the above settings, we also acquired full resolution, low frame rate video at 0.1 frames per second. Given a circular illuminated area of 0.9 cm radius, the total laser power of 50 mW is distributed over 2.5 cm 2 , yielding an average optical power of ~20 mW/cm 2 and 2 mJ/cm 2 per image frame. The resulting images have a contrast of about 10. In a commercial real-time lice counter, the optical resolution must be a compromise between the required area to be covered per frame and the size of the details to be studied. A resolution of 0.2 mm per pixel gives 15 × 15 pixels for typical lice of 3 mm length, and this should be adequate for detecting the lice by image analysis. Efforts must also be made to select an appropriate laser pulse frequency and peak power effect, since the salmon can swim at a speed of 1 m/s. There will be a need to capture images in a stop-motion-like manner to avoid blurring. To minimize the movement of the fish during image acquisition, we need a pulsed laser with a pulse width of about 0.5 ms and a repetition rate of 50 Hz. We can then image the fish in slices of 2 cm by 30 cm, as the fish swims by. Each pulse must then have a peak effect of 240 W to obtain the optical energy of 2 mJ/cm 2 per slice. To image the required area of 2 cm x 30 cm with a resolution of 0.2 mm, a camera chip with at least 100 × 1500 pixels is required. Further research is still needed to determine the minimum laser power required and the actual configuration of the laser or laser set and camera for the best performance in a real-time lice counter.
4 Conclusion
L. salmonis produces a fluorescence signal when excited with green light (532 nm). The detailed images support that the fluorescence originates from chitin, which is part of both the exoskeleton and the digestive and reproductive systems. We have shown that a simple camera system, such as the one we set up in the lab, can be used to study the anatomy of live lice without using dyes such as Congo red as a fluorescence marker. The way forward lies in performing tests and experimental work on a larger scale, including measurements in a seawater tank with live lice and live salmon. In this way we can confirm that the imaging principle is scalable and can be realized as a live display and real-time count of salmon lice in a salmon farm. Further research will include investigation of the fluorescence of other particles in the water column that can interfere with image analysis, including callanus and fluorescent feed particles from salmon farms.
Conflict of interest
The authors declare that there is no conflict of interest.
Acknowledgements
This work was carried out in collaboration with Steinsvik (now ScaleAQ, https://scaleaq.no/ ), who contributed funding, salmon and live lice for most of the experimental work. Live lice were provided by Anna Solvang Båtnes from NTNU.
For more information or a no-obligation quote, please contact us, you can also call +31 75 2022963