Lecture Notes on ACCO Guidelines and Specimen Paper lect 20

ACCO Guidelines for Facial Identification

• Overview of ACCO guidelines for facial identification.

• These guidelines outline key techniques, providing an overview of each.

• Facial composites are the most prevalent technique, used more often than facial reconstruction in police investigations.

• Distinction between sketches by a sketch artist and computerized artist impressions (facial composite construction).

• Sketches involve a sketch artist drawing a face based on a witness's description, with modifications based on feedback.

• Computerized artist impressions refer to facial composite construction.

Post Mortem Interpretation

• Post mortem interpretation involves enhancing images of deceased individuals for public circulation to aid identification.

• This technique is used when there is light damage to the soft tissues of the face.

• Forensic artists typically perform aesthetic enhancements.

• Example shown: image of the deceased on the left, post mortem interpretation on the right.

Facial Reconstructions

• Facial reconstructions are employed when there's severe damage to the soft tissue, making the face unrecognizable.

• The process involves working with the skull to rebuild the soft tissues.

• Approaches:

  • Two-dimensional reconstruction:

    • A photograph of the skull is taken.

    • Soft tissues are built up using software like Photoshop.

    • The artist inserts eyes and musculature, then adds skin, hair, etc.

    • Artistic license is used, but skull shape provides strong indicators (e.g., ear shape, face width).

    • Example: Image produced by the forensic art team at the University of Dundee (led by Sue Black).

  • Three-dimensional reconstruction:

    • Traditionally involves creating a replica of the skull and placing depth markers to indicate tissue depth.

    • Clay is used to build up the soft tissues on the skull replica.

    • This technique is referred to as the "sooty pin technique," dating back to Victorian times.

    • Glass eyes are inserted, muscles are built up, and a final layer of clay is applied for the skin.

    • The face is then painted, and hair is added.

    • The resulting representation can be rotated 360 degrees.

  • Modern approach:

    • CT scans are used to create a 3D model of the skull.

    • Depth markers are placed in 3D software.

    • Forensic artists use haptic feedback devices (e.g., a 3D joystick or pen) to build up soft tissues.

    • Haptic devices provide resistance and feedback based on the shape of the skull.

Age Progression of Children

• Involves generating images to show how a missing child might look at an older age.

• Example: Case of Patrice, who went missing at age two in the 1980s.

• An age-progressed version shows what Patrice might look like at age 28.

• Forensic artists often use images of genetic relatives (e.g., siblings) to inform the age progression.

• Requirements for age progression:

  • The child must be at least two years old when missing.

  • The child must have been missing for at least two years.

  • If the child is younger than two, images of siblings may be used due to rapid facial development.

Image Comparison

• Image comparison, formerly known as facial mapping, involves comparing two photographs to determine if they are of the same individual.

• The technique involved overlapping images in software like Photoshop and assessing how well features align.

• Facial mapping has proven unreliable.

• Modern techniques involve using face recognition software for image comparison.

• Important to distinguish between facial reconstruction and facial composite construction; they are different techniques.

Specimen Paper: Facial Identification Question

• The first question relates to the ACPO guidelines on facial ID.

• Part Ai:

  • Question: If a witness is unable to recall facial features but might recognize the suspect, what type of facial composite system should be used?

  • Answer: A holistic facial composite system such as EPIT6.

  • The EPIT6 system relies on recognition of similarities between whole faces rather than memory recall.

  • This distinction is mentioned in the ACPO guidelines under the plate Prophecy Assessment.

• Next question:

  • Relates to the acronym ADVOCATE from the terminal guidelines.

  • Requires stating what each letter in the acronym stands for.

• Last part:

  • Asks about the minimum training standards for a facial imaging officer who produces facial composites.

  • Details are in the APCO guidelines for facial ID, specifically section 3.1.

  • Minimum standards include relevant training and assessment and achieving national occupational standards set by Skills of Justice.

  • Training must cover appropriate interview techniques, production of facial images, and evidence handling preparation.

• Facial composites are rarely used as evidence in court but are a tool for locating suspects.

• Images are circulated internally within the police or to the media to trigger public memory.

Specimen Paper: Statistical Question

• Archaeological study: Cranial capacities of seven adult Neanderthal skulls were measured.

• Ten adult skulls were selected at random from a collection of modern-day homo sapien skeletal remains.

• The mean cranial capacity of these skulls was 1300 \, cm^3, and the standard deviation was 200 \, cm^3.

• Perform a two-sample t-test which involves two independent samples.

• Two-tailed test.

• Formula given: Standard Error = \sqrt{\frac{S1^2}{N1} + \frac{S2^2}{N2}}

  • Where N1 and N2 refer to the number of samples for Neanderthals and Homo sapiens, respectively.

  • S_1 is the standard deviation for Homo sapiens.

  • S_2 would be the standard deviation for Neanderthals.

• The formula provides an estimate of the pooled standard deviation.

• Scaling such that the larger the sample size, the more confident we are about the results.

• The standard error decreases as the sample size increases.

• Degrees of freedom: Total number of samples minus two.

  • In this case, 10 Homo sapiens and 7 Neanderthals, so the degrees of freedom is 15.

• Use the number 15 and the significance level (5%) to look at the statistical table.

• The value taken from the table is 2.093.

• Calculate: Difference between the means.

• T statistic: t = \frac{\bar{X1} - \bar{X2}}{SE}

• If the interval includes zero, the statistical result is not significant.

• Calculate the relevant numbers:

  • The mean value for cranial capacity for the Neanderthals is 1531.4 \, cm^3.

  • Corresponding standard deviation is 430.2.

  • When we calculate the standard deviation because we have a small sample size, our scaling parameter is one over the number of samples minus one.

• The mean and the standard deviation for the homo sapiens is given, but it's not given for the Neanderthals.

• Difference between the means is 231.4.

• Then we calculate the standard error.

• Plug numbers into the standard error formula.

• Then the next step is to look up T value for degrees of freedom for the problem.

• We said the degrees of freedom is 15.

• Looking for 15, we've got 15 degrees of freedom and that means that for 5% significance the value of T is 2.131.

• SE = 154 and t = 2.1.

• If we do 231 \pm (154 \cdot 2.1), we get the range of -92 to 555.

• In this particular case, there was no statistically significant difference between cranial capacity of Neanderthals and homo spiens.

ACCO Guidelines for Facial Identification

• Overview of ACCO guidelines for facial identification.

• These guidelines outline key techniques, providing an overview of each.

• Facial composites are the most prevalent technique, used more often than facial reconstruction in police investigations.

• Distinction between sketches by a sketch artist and computerized artist impressions (facial composite construction).
  -Sketches involve a sketch artist drawing a face based on a witness's description, with modifications based on feedback. These sketches rely heavily on the artist's skill in interpreting and translating the witness's memory into a visual representation. The process is iterative, often involving multiple revisions to refine the likeness. Sketches are advantageous for their flexibility and the ability to capture subtle details that might be missed by less nuanced methods.
  -Computerized artist impressions refer to facial composite construction, which utilizes specialized software to assemble a face from a database of facial features. This method allows for a systematic approach to creating a composite, where features can be swapped and adjusted to match the witness's recollection. The advantages of computerized composites include speed and the ability to easily modify and share the images. However, the quality of the composite is heavily reliant on the breadth and realism of the feature database.

Post Mortem Interpretation

• Post mortem interpretation involves enhancing images of deceased individuals for public circulation to aid identification. This is particularly useful when the body is unidentified and traditional identification methods have been exhausted. The goal is to make the deceased recognizable to someone who knew them, which can help in solving missing person cases or identifying individuals who have died away from home.

• This technique is used when there is light damage to the soft tissues of the face. Minor abrasions, discolorations, or distortions can be digitally corrected to present a more lifelike appearance.

• Forensic artists typically perform aesthetic enhancements, which may include removing blemishes, correcting skin tone, and adjusting facial features to match pre-death photographs or descriptions.

• Example shown: image of the deceased on the left, post mortem interpretation on the right.

Facial Reconstructions

• Facial reconstructions are employed when there's severe damage to the soft tissue, making the face unrecognizable. This technique is also used when the remains are skeletonized, and no soft tissue remains. The reconstruction aims to approximate the individual's appearance in life, based on the underlying skeletal structure.

• The process involves working with the skull to rebuild the soft tissues. This requires a deep understanding of facial anatomy, including the average tissue depths at various points on the skull.

• Approaches:

  • Two-dimensional reconstruction:

  • A photograph of the skull is taken. This is often a standardized, frontal view to maintain consistency.

    • Soft tissues are built up using software like Photoshop. The artist uses digital brushes and layers to add musculature, fat, and skin.

    • The artist inserts eyes and musculature, then adds skin, hair, etc. Eye placement and size are crucial for achieving a realistic likeness. Reconstructing the musculature helps to define the shape of the face and mouth.

    • Artistic license is used, but skull shape provides strong indicators (e.g., ear shape, face width). While artistic interpretation is necessary, the reconstruction is grounded in scientific principles and anatomical knowledge. The shape of the skull provides essential clues about the individual's facial features.

    • Example: Image produced by the forensic art team at the University of Dundee (led by Sue Black).

  • Three-dimensional reconstruction:

  • Traditionally involves creating a replica of the skull and placing depth markers to indicate tissue depth. The depth markers are based on statistical data of average tissue thicknesses for different populations, sex, and age groups.

    • Clay is used to build up the soft tissues on the skull replica. The clay is applied in layers, following the contours and landmarks of the skull.

    • This technique is referred to as the "sooty pin technique," dating back to Victorian times. The pins are inserted into the skull at measured depths to guide the application of the clay.

    • Glass eyes are inserted, muscles are built up, and a final layer of clay is applied for the skin. The glass eyes are carefully positioned to match the eye sockets. The muscles are reconstructed to define facial expressions and contours.

    • The face is then painted, and hair is added. The painting helps to bring the face to life, adding realism and detail. The hair is styled to match the estimated age and cultural background of the individual.

    • The resulting representation can be rotated 360 degrees. This allows for a comprehensive view of the reconstructed face from all angles.

  • Modern approach:

  • CT scans are used to create a 3D model of the skull. This digital model allows for precise measurements and manipulation without damaging the original skull.

    • Depth markers are placed in 3D software. The software uses algorithms to calculate tissue depths based on the individual's demographic information.

    • Forensic artists use haptic feedback devices (e.g., a 3D joystick or pen) to build up soft tissues. These devices allow the artist to feel the contours of the skull and simulate the resistance of soft tissue.

    • Haptic devices provide resistance and feedback based on the shape of the skull. This tactile feedback enhances the accuracy and realism of the reconstruction.

Age Progression of Children

• Involves generating images to show how a missing child might look at an older age. This is a crucial tool for long-term missing person cases, as it helps to keep the child's image relevant and recognizable to the public.

• Example: Case of Patrice, who went missing at age two in the 1980s.

• An age-progressed version shows what Patrice might look like at age 28. This provides a more contemporary image for potential recognition.

• Forensic artists often use images of genetic relatives (e.g., siblings) to inform the age progression. The artist analyzes facial similarities and projects how the child's features might evolve over time, based on the relatives' appearances.

• Requirements for age progression:

  • The child must be at least two years old when missing. This ensures that there is enough facial structure development to make a reliable prediction.

  • The child must have been missing for at least two years. This is to justify the need for an updated image that reflects the passage of time.

  • If the child is younger than two, images of siblings may be used due to rapid facial development. The facial features of very young children change rapidly, making it difficult to predict their future appearance without familial references.

Image Comparison

• Image comparison, formerly known as facial mapping, involves comparing two photographs to determine if they are of the same individual. This is a common task in forensic investigations, used to verify identities in surveillance footage, ID photos, and other visual evidence.

• The technique involved overlapping images in software like Photoshop and assessing how well features align. This older method relied heavily on manual alignment and subjective judgment.

• Facial mapping has proven unreliable due to its subjective nature and lack of scientific rigor. The results were often influenced by the examiner's biases and the quality of the images.

• Modern techniques involve using face recognition software for image comparison. These systems use algorithms to analyze facial features and compare them quantitatively.

• Important to distinguish between facial reconstruction and facial composite construction; they are different techniques. Facial reconstruction aims to recreate the likeness of an unidentified individual from their skeletal remains, while facial composite construction aims to create an image of a suspect based on eyewitness descriptions.

Specimen Paper: Facial Identification Question

• The first question relates to the ACPO guidelines on facial ID.

• Part Ai:

  • Question: If a witness is unable to recall facial features but might recognize the suspect, what type of facial composite system should be used?

  • Answer: A holistic facial composite system such as EPIT6. Holistic systems focus on the overall impression of the face rather than individual features.

  • The EPIT6 system relies on recognition of similarities between whole faces rather than memory recall. It presents a series of faces and asks the witness to select the one that looks most familiar.

  • This distinction is mentioned in the ACPO guidelines under the plate Prophecy Assessment. The Prophecy Assessment is a method of evaluating the reliability of eyewitness testimony by assessing the witness's confidence and consistency.

• Next question:

  • Relates to the acronym ADVOCATE from the terminal guidelines. The ADVOCATE acronym is a mnemonic used to ensure that all relevant information is gathered during an eyewitness interview.

  • Requires stating what each letter in the acronym stands for.

• Last part:

  • Asks about the minimum training standards for a facial imaging officer who produces facial composites. These standards ensure that officers are competent and qualified to perform facial imaging tasks.

  • Details are in the APCO guidelines for facial ID, specifically section 3.1. The guidelines outline the specific knowledge, skills, and abilities required for facial imaging officers.

  • Minimum standards include relevant training and assessment and achieving national occupational standards set by Skills of Justice. These standards cover various aspects of facial imaging, including interview techniques, image production, and evidence handling.

  • Training must cover appropriate interview techniques, production of facial images, and evidence handling preparation. Proper interview techniques are essential for obtaining accurate and reliable information from witnesses. Training in image production ensures that officers can create high-quality facial composites. Knowledge of evidence handling is crucial for maintaining the integrity of the evidence.

• Facial composites are rarely used as evidence in court but are a tool for locating suspects. They are more often used as an investigative aid rather than direct evidence.

• Images are circulated internally within the police or to the media to trigger public memory. This helps to generate leads and potential identifications from members of the public.

Specimen Paper: Statistical Question

• Archaeological study: Cranial capacities of seven adult Neanderthal skulls were measured. This is a common method for estimating brain size in extinct hominins.

• Ten adult skulls were selected at random from a collection of modern-day homo sapien skeletal remains. Random selection helps to ensure that the sample is representative of the population.

• The mean cranial capacity of these skulls was 1300 \, cm^3, and the standard deviation was 200 \, cm^3. The standard deviation measures the amount of variation or dispersion in the dataset.

• Perform a two-sample t-test which involves two independent samples. The t-test is used to determine if there is a statistically significant difference between the means of two groups.

• Two-tailed test. A two-tailed test is used when we are interested in detecting differences in either direction (i.e., whether one group is significantly greater or less than the other).

• Formula given: Standard Error = \sqrt{\frac{S1^2}{N1} + \frac{S2^2}{N2}}

  • Where N1 and N2 refer to the number of samples for Neanderthals and Homo sapiens, respectively.

  • S_1 is the standard deviation for Homo sapiens.

  • S_2 would be the standard deviation for Neanderthals.

• The formula provides an estimate of the pooled standard deviation. The pooled standard deviation is a weighted average of the standard deviations of the two groups.

• Scaling such that the larger the sample size, the more confident we are about the results. Larger sample sizes provide more statistical power, making it easier to detect significant differences.

• The standard error decreases as the sample size increases. The standard error is a measure of the precision of the sample mean.

• Degrees of freedom: Total number of samples minus two.

  • In this case, 10 Homo sapiens and 7 Neanderthals, so the degrees of freedom is 15.

• Use the number 15 and the significance level (5%) to look at the statistical table. The statistical table provides critical values for the t-distribution based on the degrees of freedom and significance level.

• The value taken from the table is 2.093. This is the critical value that determines the threshold for statistical significance.

• Calculate: Difference between the means.

• T statistic: t = \frac{\bar{X1} - \bar{X2}}{SE}

• If the interval includes zero, the statistical result is not significant. This means that there is no statistically significant difference between the means of the two groups.

• Calculate the relevant numbers:

  • The mean value for cranial capacity for the Neanderthals is 1531.4 \, cm^3.

  • Corresponding standard deviation is 430.2.

  • When we calculate the standard deviation because we have a small sample size, our scaling parameter is one over the number of samples minus one. This is known as Bessel's correction and provides an unbiased estimate of the population standard deviation.

• The mean and the standard deviation for the homo sapiens is given, but it's not given for the Neanderthals.

• Difference between the means is 231.4.

• Then we calculate the standard error.

• Plug numbers into the standard error formula.

• Then the next step is to look up T value for degrees of freedom for the problem.

• We said the degrees of freedom is 15.

• Looking for 15, we've got 15 degrees of freedom and that means that for 5% significance the value of T is 2.131.

• SE = 154 and t = 2.1.

• If we do 231 \pm (154 \cdot 2.1), we get the range of -92 to 555.

• In this particular case, there was no statistically significant difference between cranial capacity of Neanderthals and homo spiens.