fix: fix all strange charaters and add image for E4E 2017 Summer REU

_posts/2015-09-04-summer-reu.html

@@ -28,7 +28,7 @@
   school: University of California, San Diego
   align: alignleft
   photo: /assets/2015-09-04-sam.png
-  bio: Samuel is a senior currently pursuing a bachelor's in Electrical Engineering with a focus in computer design. He then plans on pursuing a Master’s in intelligent systems through the BSMS program at UCSD. This summer he worked on the Elephant monitoring project, as well as the initial research into the Rhino truck detection projects. In his free-time he enjoys working on the flight code for his arduino-controlled quadcopter.
+  bio: Samuel is a senior currently pursuing a bachelor's in Electrical Engineering with a focus in computer design. He then plans on pursuing a Master€™s in intelligent systems through the BSMS program at UCSD. This summer he worked on the Elephant monitoring project, as well as the initial research into the Rhino truck detection projects. In his free-time he enjoys working on the flight code for his arduino-controlled quadcopter.
 - name: Annie Christy
   school: Harvey Mudd College, Claremont, CA
   align: alignright

_posts/2015-09-15-summer-reu-bird-nest-tracker.md

@@ -31,7 +31,7 @@ The system is composed of two cameras used to record video data gathered by a SB
 
 [![]({{'/assets/2015-09-15-birdnest-images.png' | resize: '573x231'}})]({{'/assets/2015-09-15-birdnest-images.png' | absolute_url}})
 
-The two cameras have very different resolutions and fields of view, and their positions on the aerial platform will be offset slightly. This means that at the same instance in time, each camera will capture a similar but different recording of what is happening.  Sensor fusion and image processing can be used to skew, crop and rotate one of the videos so that we obtain the same image from both cameras, giving us useful data to analyze.
+The two cameras have very different resolutions and fields of view, and their positions on the aerial platform will be offset slightly. This means that at the same instance in time, each camera will capture a similar but different recording of what is happening.  Sensor fusion and image processing can be used to skew, crop and rotate one of the videos so that we obtain the same image from both cameras, giving us useful data to analyze.
 
 **Goals:**
 

_posts/2015-10-15-summer-airborne-radio-collar-tracking.md

@@ -23,7 +23,7 @@ Many wildlife ecology studies look at population size, location, and density to
 <figcaption><em>Cyclura ricordi</em> (Ricord's ground iguana) hatchling with radio collar attached, Dominican Republic, September 2015. Photo Credit: Dr. John Iverson, Earlham College</figcaption>
 </figure>
 
-For the past couple years, Engineers for Exploration has collaborated with researchers from the San Diego Zoo. In particular, we have been working with Dr. Stesha Pasachnik, who studies iguana populations in the Caribbean. We are currently developing a platform to track radio collars placed on iguana hatchlings in the Dominican Republic. Recently, a team of engineers from E4E deployed this platform into the field, once in early July, and again in mid September. These first field trials of the Radio Collar Tracker platform provided a proof-of-concept field demonstration that proved the feasibility of autonomously locating radio collars using low-cost aerial platforms.
+For the past couple years, Engineers for Exploration has collaborated with researchers from the San Diego Zoo. In particular, we have been working with Dr. Stesha Pasachnik, who studies iguana populations in the Caribbean. We are currently developing a platform to track radio collars placed on iguana hatchlings in the Dominican Republic. Recently, a team of engineers from E4E deployed this platform into the field, once in early July, and again in mid September. These first field trials of the Radio Collar Tracker platform provided a proof-of-concept field demonstration that proved the feasibility of autonomously locating radio collars using low-cost aerial platforms.
 
 <figure>
 <a href="{{'/assets/2015-10-15-dom-rep-takeoff.jpg' | absolute_url}}"><img src="{{'/assets/2015-10-15-dom-rep-takeoff.jpg' | resize: '1024x768'}}"></a>

_posts/2016-08-11-2016-summer-reu.md

@@ -102,7 +102,7 @@ Irina Tolkova is a Applied Math and Computer Science senior at the University of
 
 
 [![2016 Tara Tripp](/assets/2016-08-11-2016-summer-reu_2016_tara.jpg)](/assets/2016-08-11-2016-summer-reu_2016_tara.jpg)
-**Tara Tripp, University of Illinois, Urbana–Champaign**
+**Tara Tripp, University of Illinois, Urbana€“Champaign**
 
 Tara is a third year computer engineering student at UIUC. Over the summer she worked on developing a terrain avoidance system for multirotor drones using rangefinder and terrain elevation data in order to improve aerial 3D mapping. In her free time, Tara enjoys reading, playing video games, and working with her robotics team.
 
@@ -113,7 +113,7 @@ Tara is a third year computer engineering student at UIUC. Over the summer she w
 [![2016 Daniel Webber](/assets/2016-08-11-2016-summer-reu_2016_daniel-300x200.jpg)](/assets/2016-08-11-2016-summer-reu_2016_daniel.jpg)
 **Daniel Webber, Santa Clara University**
 
-Daniel is a rising sophomore currently pursuing a bachelor's degree in Electrical Engineering. This summer he worked on the radio collar tracker platform, improving the receiver’s hardware to make the system more sensitive to increase the detection range. In his free-time he enjoys flying quadcopters, cycling and photography. 
+Daniel is a rising sophomore currently pursuing a bachelor's degree in Electrical Engineering. This summer he worked on the radio collar tracker platform, improving the receiver€™s hardware to make the system more sensitive to increase the detection range. In his free-time he enjoys flying quadcopters, cycling and photography. 
 
 
 
@@ -153,7 +153,7 @@ Quentin is a PhD student in Computer Science at UC San Diego. He moved from Fran
 
 [![Nathan Hui](/assets/REU2015_Nathan-229x300.png)](/assets/REU2015_Nathan-229x300.png)**Nathan Hui**
 
-Nathan Hui is a 5th year Electrical Engineering Student at UC San Diego. He is currently the Project Lead for the Radio Collar Tracker, a collaboration between UC San Diego and the San Diego Zoo. Hui has been a member of Engineers for Exploration since 2014, working on projects including the Aerial Balloon Platform. In addition to being a student and undergraduate researcher, Hui is an Outings Guide with UC San Diego’s Outback Adventures, guiding students on trips throughout the Southwestern United States and Western Mexico.
+Nathan Hui is a 5th year Electrical Engineering Student at UC San Diego. He is currently the Project Lead for the Radio Collar Tracker, a collaboration between UC San Diego and the San Diego Zoo. Hui has been a member of Engineers for Exploration since 2014, working on projects including the Aerial Balloon Platform. In addition to being a student and undergraduate researcher, Hui is an Outings Guide with UC San Diego€™s Outback Adventures, guiding students on trips throughout the Southwestern United States and Western Mexico.
 
 
 

_posts/2016-08-29-summer-2016-airborne-radio-collar-tracking.md

@@ -19,7 +19,7 @@ tags:
 - summer
 ---
 
-The Radio Collar Tracker project is a collaboration between the San Diego Zoo’s Beckman Institute for Conservation Research (ICR) and UC San Diego’s Engineers For Exploration. We have been working with the ICR and its collaborators for the past four years to develop a drone to conduct radio collar tracking missions. Ultimately, the goal of this project is to develop a drone capable of autonomously tracking radio collared animals, particularly through environments that are difficult to access. To accomplish this, our system flies a search grid over a candidate area while recording a large swath of radio spectrum using a software defined radio. Once the drone returns from the flight, we use digital signal processing to extract the frequencies that the radio collars are transmitting on, and use a statistical model to determine where the collars we heard are. This data is then saved to a GIS (Geographical Information System) file, which ecologists and biologists can then use in their research.
+The Radio Collar Tracker project is a collaboration between the San Diego Zoo€™s Beckman Institute for Conservation Research (ICR) and UC San Diego€™s Engineers For Exploration. We have been working with the ICR and its collaborators for the past four years to develop a drone to conduct radio collar tracking missions. Ultimately, the goal of this project is to develop a drone capable of autonomously tracking radio collared animals, particularly through environments that are difficult to access. To accomplish this, our system flies a search grid over a candidate area while recording a large swath of radio spectrum using a software defined radio. Once the drone returns from the flight, we use digital signal processing to extract the frequencies that the radio collars are transmitting on, and use a statistical model to determine where the collars we heard are. This data is then saved to a GIS (Geographical Information System) file, which ecologists and biologists can then use in their research.
 
 For the past couple of years, the Radio Collar Tracker has been further developed. In the summer of 2015, a proof of concept was built and deployed in the Dominican Republic. While there were problems, it demonstrated the potential for the system to be effective. Since then, some of the post-processing and workflow have been improved to make the system more effective and help the researchers better visualize where the radio collar is located. In the spring of this year, the system was deployed in the Cayman Islands, but some of the same problems persisted. One key problem of the system is the detection range. Currently the system has a detection range of about 30 meters. While this is okay, to make the system really effective the range has to be improved. For the summer the goal was to improve the range from a hardware perspective.
 
@@ -31,8 +31,8 @@ The new design consisted of the same components: antenna and a software defined
 
 [caption id="attachment_4165" align="aligncenter" width="640"]![New Radio Collar Tracker RF Equipment](/assets/2016-08-29-summer-2016-airborne-radio-collar-tracking_IMG_0786-1024x768.jpg) An example of the receiver with the SDR, amplifier, and antenna[/caption]
 
-Although we didn’t test the system on a field deployment like last summer, we still tried to not always test in the lab. Unfortunately due increasing restriction on flying UAS (Unmanned Aerial Systems) for research we were unable to test the system locally. To get the best idea of the improvement we tested our system on the ground. Although this test would not give us an absolute gauge of improvement, it can show it in a relative manner. From the initial results we have been able to demonstrate that we have almost doubled the detection range compared to the system that was on previous deployments.
+Although we didn€™t test the system on a field deployment like last summer, we still tried to not always test in the lab. Unfortunately due increasing restriction on flying UAS (Unmanned Aerial Systems) for research we were unable to test the system locally. To get the best idea of the improvement we tested our system on the ground. Although this test would not give us an absolute gauge of improvement, it can show it in a relative manner. From the initial results we have been able to demonstrate that we have almost doubled the detection range compared to the system that was on previous deployments.
 
 Ultimately this summer we have designed the receiver to become more sensitive and we are excited to see the results on the next field deployment.
 
-__—_ by Daniel Webber and Nathan Hui_
+__€”_ by Daniel Webber and Nathan Hui_

_posts/2016-09-03-machine-learning-for-humpback-whale-social-call-classification-2016-reu.md

@@ -26,7 +26,7 @@ Our research this summer involved experimenting with various methods of data pro
 [![whale_spectrogram_original](/assets/2016-09-03-machine-learning-for-humpback-whale-social-call-classification-2016-reu_whale_spectrogram_original-1024x280.png)](/assets/2016-09-03-machine-learning-for-humpback-whale-social-call-classification-2016-reu_whale_spectrogram_original.png)
 
 
-Above are ten spectrograms – a visual representation of acoustic data that shows the intensity of sound across different frequencies through time. The whale calls are boxed and labeled.
+Above are ten spectrograms €“ a visual representation of acoustic data that shows the intensity of sound across different frequencies through time. The whale calls are boxed and labeled.
 
 
 
@@ -61,4 +61,4 @@ Additionally, the noise across different data sets varied causing individual dat
 
 
 
-_— by Lisa Bauer, Irina Tolkova, Hongyi Zhao, and Antonella Wilby_
+_€” by Lisa Bauer, Irina Tolkova, Hongyi Zhao, and Antonella Wilby_

_posts/2016-09-20-spherecam-for-marine-monitoring-2016-reu.md

@@ -50,7 +50,7 @@ While this is not the most exciting part of the project, it is an import part of
 
 
 
-Triggering was the biggest technical challenge this summer. The Vaquita uses echolocation clicks that are at around 139kHz (For reference Humans can hear from 20Hz – 20kHz). The Nyquist theorem also stipulates that to record an analog signal accurately, like a sound, digitally we need to sample at least twice the rate of the sound we want to record. Because of this, we need to sample the audio signal at about 300kHz to be able to detect the 139kHz tone, or in other words, the computer has to read the microphone signal and record a value 300,000 times per second. 
+Triggering was the biggest technical challenge this summer. The Vaquita uses echolocation clicks that are at around 139kHz (For reference Humans can hear from 20Hz €“ 20kHz). The Nyquist theorem also stipulates that to record an analog signal accurately, like a sound, digitally we need to sample at least twice the rate of the sound we want to record. Because of this, we need to sample the audio signal at about 300kHz to be able to detect the 139kHz tone, or in other words, the computer has to read the microphone signal and record a value 300,000 times per second. 
 
 ![triggering](/assets/2016-09-20-spherecam-for-marine-monitoring-2016-reu_triggering-300x229.png)
 
@@ -65,7 +65,7 @@ How long does it the computer take to read each value? How much time does that l
 
 ![edison_adc](/assets/2016-09-20-spherecam-for-marine-monitoring-2016-reu_edison_adc-300x300.jpg)
 
-**Edison ADC breakout:** Made by Sparkfun this board provides a “built-in” ability for the Edison to take readings from a microphone using an Analog to Digital Converter (ADC). Because Linux is not a real-time system and there is an inherent latency with accessing the ports on the Edison this was able to sample at a maximum rate of about 9,000 Hz. Not good enough!
+**Edison ADC breakout:** Made by Sparkfun this board provides a €œbuilt-in€ ability for the Edison to take readings from a microphone using an Analog to Digital Converter (ADC). Because Linux is not a real-time system and there is an inherent latency with accessing the ports on the Edison this was able to sample at a maximum rate of about 9,000 Hz. Not good enough!
 
 
 ![spherecam-edisons](/assets/2016-09-20-spherecam-for-marine-monitoring-2016-reu_spherecam-edisons-286x300.jpg)
@@ -111,4 +111,4 @@ Ethan makes some repairs in the field
 
 In early September, we deployed the SphereCam in the Vaquita Refuge. It will remain in the refuge until late September, at which point we will return to swap the batteries, grab the data, and clean the domes of biofouling. Check back later for more updates on what the SphereCam sees!
 
-_— by Ethan Slattery, Andrew Hostler and Antonella Wilby_
+_€” by Ethan Slattery, Andrew Hostler and Antonella Wilby_

_posts/2016-09-30-3d-visualization-rig-for-imaging-underwater-lemur-graveyards-2016-reu.md

@@ -24,4 +24,4 @@ A mysterious graveyard of giant lemur fossils has been found in the underwater A
 Our solution is a small, easily maneuverable diving rig with cameras controlled by the diver via an iPad. The cameras will be mounted inside clear acrylic domed housings, and a minnowboard will connect them to the iPad. Our project goal is to produce reliable and good quality images from the site, so the place and all the other elements can be studied and understood by scientists and explorers. 
 
 
-_— by Rachel Herrera, Daniel Knapp, Kaue Zoia, Miguel de Villa and Antonella Wilby_
+_€” by Rachel Herrera, Daniel Knapp, Kaue Zoia, Miguel de Villa and Antonella Wilby_