The histogram is the topic for the week in the Boost Your Photography: 52 Weeks Challenge. (Join the Google+ Community to share your weekly photographs and receive feedback.) This article seeks to demystify some of the questions and confusion that generally arise when discussing histograms.
Histogram: a histogram is a bar graph of the frequency distribution of tones in a photograph, ranging from pure black (0) to pure white (255). The higher the bar graph for a certain segment, the more that particular tone is represented in the photograph. There are two types of histograms: one that combines three different bar graphs for each of the three different color channels (blue, red, and green) and one that provides one bar graph for the entire photograph. We will stick to talking about the single, overall histogram for now, but you can see both versions below.
Clipping: clipping is a term used to refer to elements of the photograph that are recorded as either pure black or pure white. Once a section of a photograph has been "clipped," it becomes difficult to recover any detail from that particular part of the image, even with post-processing. (Think: a black silhouette or an all white window in an indoor photograph.) An ideal histogram includes nearly the entire range of tones from 0 to 255 with a minimum amount of clipping. In the example above, you can see that the histogram values nearly approach pure black (0) but not quite. This means that there is no clipping of black values, and you can see still subtle variations in the shadows of the framing leaves.
The Blinkies: the blinkies (or, the dreaded blinkies) are a warning that you can enable on many cameras to alert you to clipping in your photographs. After taking a photograph, you can review it on your camera screen, and you will see blinking sections of your photograph that indicate that those sections have been clipped. (In my Canon instructional manual, the technical term is "highlight alert," and it blinks only for overexposed areas, not underexposed ones.) In the example above, I used Levels to artificially clip many of the near-black values and change them to pure black (upper histogram shows levels). Now you can see that there is a greater frequency of values on the histogram that are at or near pure black (lower histogram). Detail has been lost in these areas.
A general histogram of a well-exposed and well-balanced image will have a familiar, bell-curve shaped look. The edges of the curve should reach just towards the two edges but not touch (no clipping), as in the photograph above. This is an idealized version of what an appropriate exposure should look like for a photograph. Keep in mind, however, that every photographic situation is different, and a bell curve will not always represent what you are looking to convey.
A high-key histogram is one where much of the histogram is squished up to the right-hand side of the graph. Large sections of the photograph will be clipped and overexposed (rendered as pure white). The photograph above shows an extreme example of a high-key photograph and matching histogram.
A low-key histogram is one where much of the histogram is squished up the left-hand side of the graph. Large sections of the photograph will be clipped and underexposed (rendered as pure black). The photograph above shows an extreme example of a low-key photograph and matching histogram.
A low contrast histogram is one where the histogram is confined to the middle values on the graph and does not approach either the high or low ends of the histogram. Low contrast shots are common in situations with little or minimal contrast, such as a foggy or snowy day like in the example shot and histogram above. You can see the lack of pure white both in the greys of the photograph and in the absence of white values on the right-hand side of the histogram. There is also only minimal pure black values on the left-hand side of the histogram.
The graphic above compares two different photographs of the Grand Tetons, shot using exposure compensation. The top image was at 0, while the bottom was at -1 exposure. With the top image, it is clear that the clouds in the sky and the reflection are clipped (blown out), and the histogram shows that with the values on the right-hand side. The bottom image shows a more even distribution of tones across the full range. In this situation, using a negative value for exposure compensation allowed me to shift the histogram over towards the left. (Read more about exposure compensation in Explaining Exposure and Exposure Compensation.)
Other examples of when to use your histogram: if you are looking to capture dark silhouettes against a bright background, you can check your histogram to see if you have achieved pure black (a clustering of bars against the left-hand side). If you are looking to capture a subject against a bright white background, you can check your histogram to see if you have achieved pure white (a clustering of bars against the right-hand side).
How will you use your histogram? Share your thoughts or an example in the comments below.
(Looking to grow more in your photography? Consider joining the BYP 52 Weeks Google+ Community to share your weekly photograph and see what others are capturing.)
Boost Your Photography: Learn Your DSLR is now available from Amazon. Get the most out of your camera with practical advice about the technical and creative aspects of DSLR photography that will have you taking beautiful pictures right away.
Histogram Terminology
Let's start with some basic definitions, as there are several different terms commonly used when discussing histograms.Histogram: a histogram is a bar graph of the frequency distribution of tones in a photograph, ranging from pure black (0) to pure white (255). The higher the bar graph for a certain segment, the more that particular tone is represented in the photograph. There are two types of histograms: one that combines three different bar graphs for each of the three different color channels (blue, red, and green) and one that provides one bar graph for the entire photograph. We will stick to talking about the single, overall histogram for now, but you can see both versions below.
Clipping: clipping is a term used to refer to elements of the photograph that are recorded as either pure black or pure white. Once a section of a photograph has been "clipped," it becomes difficult to recover any detail from that particular part of the image, even with post-processing. (Think: a black silhouette or an all white window in an indoor photograph.) An ideal histogram includes nearly the entire range of tones from 0 to 255 with a minimum amount of clipping. In the example above, you can see that the histogram values nearly approach pure black (0) but not quite. This means that there is no clipping of black values, and you can see still subtle variations in the shadows of the framing leaves.
Example Histograms
The simple answer is that there is not a "perfect" histogram. Every photograph is different, and there may be times when you deliberately want a mostly-bright tones photograph or a mostly-dark tones photograph.A general histogram of a well-exposed and well-balanced image will have a familiar, bell-curve shaped look. The edges of the curve should reach just towards the two edges but not touch (no clipping), as in the photograph above. This is an idealized version of what an appropriate exposure should look like for a photograph. Keep in mind, however, that every photographic situation is different, and a bell curve will not always represent what you are looking to convey.
How to Use Your Histogram
Now that you have seen some extreme examples of histograms, you should have a general idea of what to expect from a given histogram and its accompanying photograph. If you are looking for a balanced photograph with a range of tones, you will want a histogram that fits the "general" example above. If your histogram is off-center, you can use exposure compensation to make your next photograph correspondingly lighter or darker as needed.The graphic above compares two different photographs of the Grand Tetons, shot using exposure compensation. The top image was at 0, while the bottom was at -1 exposure. With the top image, it is clear that the clouds in the sky and the reflection are clipped (blown out), and the histogram shows that with the values on the right-hand side. The bottom image shows a more even distribution of tones across the full range. In this situation, using a negative value for exposure compensation allowed me to shift the histogram over towards the left. (Read more about exposure compensation in Explaining Exposure and Exposure Compensation.)
Other examples of when to use your histogram: if you are looking to capture dark silhouettes against a bright background, you can check your histogram to see if you have achieved pure black (a clustering of bars against the left-hand side). If you are looking to capture a subject against a bright white background, you can check your histogram to see if you have achieved pure white (a clustering of bars against the right-hand side).
How will you use your histogram? Share your thoughts or an example in the comments below.
(Looking to grow more in your photography? Consider joining the BYP 52 Weeks Google+ Community to share your weekly photograph and see what others are capturing.)
Boost Your Photography: Learn Your DSLR is now available from Amazon. Get the most out of your camera with practical advice about the technical and creative aspects of DSLR photography that will have you taking beautiful pictures right away.
